Identifying the gaps between needs, expectations, and views of different stakeholders related to car-sharing, bike-sharing, and scooter-sharing systems

Abstract

Shared mobility, such as car-sharing, bike-sharing, and scooter-sharing services, is quickly expanding in several countries, including Italy, where it was introduced a few years ago. The benefits of this type of transportation mode have been estimated and reported by many authors. However, since a shared mobility system is a type of transportation that combines the characteristics of private vehicles and transit services, policy-makers may not know how to treat it well. Moreover, although many policies have been proposed to promote shared mobility, they still have little impact in terms of aggregated market shares in urban areas. It may be because the actual requirements of the passengers regarding shared transportation services characteristics are not well understood. Hence, it is important to understand what needs to be improved in shared mobility services. Aiming to contribute to filling this gap, two separate studies are carried out, namely the analysis of car sharing, scooter sharing, and bike sharing (separately) and the analysis of shared mobility services (as a whole, not related to a specific one). In the analysis of each shared mobility service (separately), 12 sub-criteria are compared by four different stakeholder groups (users, non-users, local authorities, and services operators) to determine their standpoints on the importance of each sub-criterion that people can consider in their decisions to use each shared mobility service. Also, in the separate analysis of each shared mobility service, each stakeholder rated the importance of specific criteria associated with their specific role. Hence, the criteria rated by government members differ from those rated by operators and users/non-users. However, users and non-users rated the same criteria in order to understand their perceptions' gaps. This study applies Multi-Actor Multi-Criteria Analysis (MAMCA) because it is an appropriate method when different stakeholders are involved. One step of the MAMCA is to determine the main criteria and weights, which is done through a perception-based analysis that was implemented by using a Bayesian Best-Worst Method (BWM). This method is chosen because it is the only one ensuring a very high quality of the computed weights while requiring a small amount of data. The latter aspect is essential because some of the shareholders are members of the government and operators, which are few in number. Other advantages of this method include the combination of weight quality, fewer inconsistencies between criteria, fewer data required to obtain highly reliable results, low equalizing bias, and average transparency of the method. Before calculating the optimal group weights by Bayesian BWM, the consistency of the interviewees’ answers was checked using the input-based approach, and acceptable ones (their obtained global input-based consistency ratio is less than the input-based consistency ratio thresholds) were considered. After eliminating pairwise comparisons with unacceptable consistency ratios, different sample sizes can be obtained and utilized for different levels of the model. Also, it is important to note that Bayesian BWM can provide much more information than the original BWM. For example, Bayesian BWM can provide the credal ranking and confidence level in the weight-directed graph. This helps to understand the importance perceived by stakeholders of one criterion over other criteria. From a methodological viewpoint, the experimental design proposed in the present work also helps to make some original contributions to the field of multicriteria analyses and Bayesian BWM applications. In order to collect the required data, nine different surveys have been designed and administered in the Turin metropolitan area in Italy. Data on operators and government members were collected through phone calls to targeted contact points, while for users and non-users, a panel maintained by a survey company was used to have a representative sample of the population in the study area (using online surveys). Survey data are used to calculate criteria and sub-criteria weights to determine how the comparative criteria are rated in terms of importance by different stakeholders of different shared mobility services. Hence, surveys provide insights into how specific individuals or groups perceive certain aspects. In those surveys administered to users and non-users of each shared mobility service, in addition to BWM-related questions, questions about their routines, daily travel views, and socio-demographic characteristics were also asked. This study helps determine the relative importance of sub-criteria and main-criteria from each stakeholder's perspective and contributes to understanding how one main-criterion/sub-criterion can be of different importance across different shared mobility services. Besides, it helps to distinguish stakeholders’ views on each sub-criterion and, more specifically, to know how different stakeholders score the importance of the comparison factors associated with their role as shared mobility service stakeholders. Based on these results, suggestions for government members and each shared mobility service operator are given to attract more users and non-users and to understand which shared mobility system is most appropriate to implement in Turin, according to users' and non-users' perceptions. Also, this study contributes to presenting scenarios to determine how to increase the use of bike-sharing and scooter-sharing services compared to car-sharing services, given their larger social benefits.

Full text

Doctoral Dissertation
Doctoral Program in Civil and Environmental Engineering (35th Cycle)
Identifying the gaps between needs,
expectations, and views of different
stakeholders related to car-sharing, bike-
sharing, and scooter-sharing systems
Ehsan Amirnazmiafshar
******
Supervisor(s):
Prof. Marco Diana
Doctoral Examination Committee:
Prof. Giovanna Acampa, Referee, Università degli Studi di Enna "Kore"
Prof. Marta Carla Bottero, Politecnico di Torino
Prof. Bruno Dalla Chiara, Politecnico di Torino
Prof. Szabolcs Duleba, Budapesti Műszaki és Gazdaságtudományi Egyetem
Prof. Mehtap Dursun Karahüseyin, Referee, Galatasaray Üniversitesi
Politecnico di Torino
February 01, 2023

Declaration
I hereby declare that, the contents and organization of this dissertation constitute my own original work
and do not compromise in any way the rights of third parties, including those relating to the security of
personal data.
Ehsan Amirnazmiafshar
Turin, February 01, 2023
* This dissertation is presented in partial fulfillment of the requirements for a Ph.D. degree in the
Graduate School of Politecnico di Torino (ScuDo).

I would like to dedicate this thesis to my loving parents

Acknowledgment
I would like to express my gratitude to my Ph.D. supervisor, Professor Marco Diana, who guided me
throughout this project.

I
Abstract
Shared mobility, such as car-sharing, bike-sharing, and scooter-sharing services, is quickly
expanding in several countries, including Italy, where it was introduced a few years ago. The
benefits of this type of transportation mode have been estimated and reported by many authors.
However, since a shared mobility system is a type of transportation that combines the
characteristics of private vehicles and transit services, policy-makers may not know how to
treat it well. Moreover, although many policies have been proposed to promote shared mobility,
they still have little impact in terms of aggregated market shares in urban areas. It may be
because the actual requirements of the passengers regarding shared transportation services
characteristics are not well understood. Hence, it is important to understand what needs to be
improved in shared mobility services.
Aiming to contribute to filling this gap, two separate studies are carried out, namely the
analysis of car sharing, scooter sharing, and bike sharing (separately) and the analysis of shared
mobility services (as a whole, not related to a specific one). In the analysis of each shared
mobility service (separately), 12 sub-criteria are compared by four different stakeholder groups
(users, non-users, local authorities, and services operators) to determine their standpoints on
the importance of each sub-criterion that people can consider in their decisions to use each
shared mobility service. Also, in the separate analysis of each shared mobility service, each
stakeholder rated the importance of specific criteria associated with their specific role. Hence,
the criteria rated by government members differ from those rated by operators and users/non-
users. However, users and non-users rated the same criteria in order to understand their
perceptions' gaps.
This study applies Multi-Actor Multi-Criteria Analysis (MAMCA) because it is an
appropriate method when different stakeholders are involved. One step of the MAMCA is to
determine the main criteria and weights, which is done through a perception-based analysis
that was implemented by using a Bayesian Best-Worst Method (BWM). This method is chosen
because it is the only one ensuring a very high quality of the computed weights while requiring
a small amount of data. The latter aspect is essential because some of the shareholders are
members of the government and operators, which are few in number. Other advantages of this
method include the combination of weight quality, fewer inconsistencies between criteria,
fewer data required to obtain highly reliable results, low equalizing bias, and average
transparency of the method.
Before calculating the optimal group weights by Bayesian BWM, the consistency of the
interviewees’ answers was checked using the input-based approach, and acceptable ones (their
obtained global input-based consistency ratio is less than the input-based consistency ratio
thresholds) were considered. After eliminating pairwise comparisons with unacceptable
consistency ratios, different sample sizes can be obtained and utilized for different levels of the
model. Also, it is important to note that Bayesian BWM can provide much more information
than the original BWM. For example, Bayesian BWM can provide the credal ranking and

II
confidence level in the weight-directed graph. This helps to understand the importance
perceived by stakeholders of one criterion over other criteria. From a methodological
viewpoint, the experimental design proposed in the present work also helps to make some
original contributions to the field of multicriteria analyses and Bayesian BWM applications.
In order to collect the required data, nine different surveys have been designed and
administered in the Turin metropolitan area in Italy. Data on operators and government
members were collected through phone calls to targeted contact points, while for users and
non-users, a panel maintained by a survey company was used to have a representative sample
of the population in the study area (using online surveys). Survey data are used to calculate
criteria and sub-criteria weights to determine how the comparative criteria are rated in terms of
importance by different stakeholders of different shared mobility services. Hence, surveys
provide insights into how specific individuals or groups perceive certain aspects. In those
surveys administered to users and non-users of each shared mobility service, in addition to
BWM-related questions, questions about their routines, daily travel views, and socio-
demographic characteristics were also asked.
This study helps determine the relative importance of sub-criteria and main-criteria
from each stakeholder's perspective and contributes to understanding how one main-
criterion/sub-criterion can be of different importance across different shared mobility services.
Besides, it helps to distinguish stakeholders’ views on each sub-criterion and, more
specifically, to know how different stakeholders score the importance of the comparison factors
associated with their role as shared mobility service stakeholders. Based on these results,
suggestions for government members and each shared mobility service operator are given to
attract more users and non-users and to understand which shared mobility system is most
appropriate to implement in Turin, according to users' and non-users' perceptions. Also, this
study contributes to presenting scenarios to determine how to increase the use of bike-sharing
and scooter-sharing services compared to car-sharing services, given their larger social
benefits.

III
Sintesi
La mobilità condivisa, come i servizi di car-sharing, bike-sharing e sharing di monopattini
elettrici, si sta espandendo rapidamente in diversi paesi, tra cui l'Italia, dove è stata introdotta
alcuni anni fa. I vantaggi di questo tipo di modalità di trasporto sono stati stimati e riportati da
molti autori. Tuttavia, poiché un sistema di mobilità condivisa è un tipo di trasporto che
combina le caratteristiche dei veicoli privati e dei servizi di trasporto pubblico, i decisori
pubblici potrebbero non sapere bene come considerarlo. Inoltre, sebbene molte politiche siano
state proposte per promuovere la mobilità condivisa, hanno ancora scarso impatto in termini di
quote di mercato aggregate nelle aree urbane. Questo potrebbe essere dovuto al fatto che le
effettive esigenze dei passeggeri in merito alle caratteristiche dei servizi di trasporto condiviso
non sono ben comprese. Pertanto, è importante capire cosa deve essere migliorato nei servizi
di mobilità condivisa.
Con l'obiettivo di contribuire a colmare questa lacuna, vengono condotti due studi
distinti, ovvero l'analisi del car sharing, sharing di monopattini elettrici e bike sharing
(separatamente) e l'analisi dei servizi di mobilità condivisa (nel loro insieme). Nell'analisi di
ciascun servizio di mobilità condivisa (separatamente), 12 sottocriteri vengono confrontati da
quattro diversi gruppi di stakeholder (utenti, non utenti, enti locali e operatori di servizi) per
determinare il loro punto di vista sull'importanza di ciascun sottocriterio che i potenziali utenti
potrebbero considerare nelle loro decisioni di utilizzare ciascun servizio di mobilità condivisa.
Inoltre, nell'analisi separata di ciascun servizio di mobilità condivisa, ogni stakeholder ha
valutato l'importanza di criteri specifici associati al proprio ruolo specifico. Pertanto, i criteri
valutati dai membri del governo differiscono da quelli valutati dagli operatori e dagli utenti/non
utenti. Tuttavia, utenti e non utenti hanno valutato gli stessi criteri per comprendere le lacune
delle loro percezioni.
Questo studio applica la Multi-Actor Multi-Criteria Analysis (MAMCA) perché è un
metodo appropriato quando sono coinvolti diversi stakeholder. Una fase del MAMCA è
determinare i criteri e i pesi principali, che viene eseguita attraverso un'analisi basata sulla
percezione che è stata implementata utilizzando un Bayesian Best-Worst Method (BWM).
Questo metodo è scelto perché è l'unico che garantisce una qualità molto elevata dei pesi
calcolati, pur richiedendo una piccola quantità di dati. Quest'ultimo aspetto è essenziale perché
alcuni dei portatori di interesse sono membri del governo e operatori, che sono pochi di numero.
Altri vantaggi di questo metodo includono la robustezza dei pesi ottenuti, meno incoerenze tra
i criteri, meno dati necessari per ottenere risultati altamente affidabili, bassa distorsione di
equalizzazione e trasparenza del metodo di calcolo.
Prima di calcolare i pesi di gruppo ottimali mediante Bayesian BWM, la coerenza delle
risposte degli intervistati è stata verificata utilizzando l'approccio basato sull'input e sono state

IV
selezionate quelle accettabili (il loro input-based consistency ratio è inferiore ad una
predeterminata soglia). Dopo aver eliminato il rischio di effettuare confronti a coppie con
rapporti di consistenza inaccettabili, è possibile ottenere e utilizzare diverse dimensioni del
campione per diversi livelli del modello. Inoltre, è importante notare che il Bayesian BWM
può fornire molte più informazioni rispetto al BWM originale. Ad esempio, il Bayesian BWM
può fornire un credal ranking e il livello di confidenza. Questo aiuta a comprendere
l'importanza percepita dalle parti interessate di un criterio rispetto ad altri criteri. Da un punto
di vista metodologico, il disegno sperimentale proposto in questo lavoro contribuisce anche a
fornire alcuni contributi originali nel campo delle analisi multicriteri e delle applicazioni del
Bayesian BWM.
Per raccogliere i dati richiesti, sono state progettate e gestite nove diverse indagini
nell'area metropolitana di Torino, in Italia. I dati su operatori e decisori pubblici sono stati
raccolti contattandoli direttamente al telefono, mentre per utenti e non utenti è stato utilizzato
un panel gestito da una società di indagine per avere un campione rappresentativo della
popolazione nell'area di studio (tramite sondaggio online) . I dati dell'indagine vengono
utilizzati per calcolare criteri e pesi dei sottocriteri, al fine di determinare in che modo i criteri
comparativi sono valutati in termini di importanza dai diversi stakeholder dei diversi servizi di
mobilità condivisa. Pertanto, le indagini forniscono informazioni su come individui o gruppi
specifici percepiscono determinati aspetti. Nelle indagini somministrate agli utenti e ai non
utenti di ciascun servizio di mobilità condivisa, oltre alle domande relative alla BWM, sono
state poste anche domande sulle loro abitudini, sugli spostamenti quotidiani e sulle loro
caratteristiche socio-demografiche.
Questo studio aiuta a determinare l'importanza relativa dei sottocriteri e dei criteri
principali secondo il punto di vista di ciascuna parte interessata e contribuisce a far
comprendere come un criterio/sottocriterio principale possa avere un'importanza diversa nei
diversi servizi di mobilità condivisa. Inoltre, aiuta a distinguere le opinioni degli stakeholder
su ciascun sottocriterio e, più specificamente, a sapere in che modo i diversi stakeholder
valutano l'importanza dei fattori di confronto maggiormente associati al loro ruolo. Sulla base
di questi risultati, vengono forniti suggerimenti ai decisori pubblici e a ciascun operatore del
servizio di mobilità condivisa per attirare più utenti e non utenti e per capire quale sistema di
mobilità condivisa è più appropriato implementare a Torino, secondo le percezioni di utenti e
non utenti. Inoltre, questo studio contribuisce a presentare scenari per determinare come
aumentare l'uso dei servizi di bike sharing e sharing di monopattini elettrici rispetto ai servizi
di car sharing, dati i loro maggiori benefici social.

V
Contents
Introduction ........................................................................................................ 1
Literature Review............................................................................................... 7
2.1 An overview of car-sharing ..................................................................... 7
2.1.1 History and trends of car-sharing systems ........................................ 7
2.1.2 Car-sharing classification .................................................................. 9
2.1.3 Interaction with other modes of transport ....................................... 12
2.1.4 Factors influencing demand for car-sharing system ....................... 15
2.1.5 Interaction effects among different factors ..................................... 39
2.1.6 Summary ......................................................................................... 42
2.2 An overview of bike-sharing ................................................................. 45
2.2.1 A brief history of bike-sharing ........................................................ 45
2.2.2 Integration of bike-sharing with other transport modes .................. 47
2.2.3 Bike and its benefits ........................................................................ 48
2.2.4 Factors affecting demand for bike .................................................. 49
2.2.5 Summary ......................................................................................... 54
2.2.6 Factors affecting demand for bike-sharing ..................................... 55
2.2.7 Summary ......................................................................................... 67
2.3 An overview of scooter-sharing ............................................................ 70
2.3.1 A brief history of e-scooter-sharing ................................................ 70
2.3.2 General advantages and disadvantages of e-scooters ..................... 70
2.3.3 E-scooter vs. other transport modes ................................................ 73
2.3.4 Factors affecting demand for e-scooters ......................................... 74
2.3.5 Summary ......................................................................................... 79
2.4 Definition of the criteria and sub-criteria that impact the demand for different
shared mobility services ..................................................................................... 83

VI
Methodology: Multi-Criteria Decision-Making Methods ............................... 87
3.1 Multi-actor multi-criteria analysis ......................................................... 89
3.1.1. Defining the problem and specifying alternatives (step 1) ............ 90
3.1.2. Stakeholder analysis (step 2) ......................................................... 90
3.1.3. Specify criteria and weights (step 3) .............................................. 90
3.1.4. Criteria, indicators, and measurement methods (step 4) ................ 90
3.1.5. Overall analysis and ranking (step 5) ............................................. 90
3.1.6. Results (step 6) ............................................................................... 91
3.1.7. Implementation (step 7) ................................................................. 91
3.2 Presentation of different MCDM methods ............................................ 91
3.2.1 Elimination and choice translating reality ...................................... 92
3.2.2 Weighted sum model ...................................................................... 92
3.2.3 Weighted product model ................................................................. 93
3.2.4 Analytic hierarchy process .............................................................. 93
3.2.5 Technique for order preference by similarities to ideal solution .... 95
3.2.6 Preference ranking organization method for enrichment evaluation97
3.2.7 Best Worst Method ......................................................................... 99
3.3 Comparative analysis and selection of the MCDM method that will be used 115
Method Implementation ................................................................................. 118
4.1. Problem definition and alternatives selection ..................................... 118
4.2 Stakeholder analysis ............................................................................ 119
4.3 Selection of criteria .............................................................................. 122
4.3.1. Analysis of perspectives of stakeholders of shared mobility services (as a
whole, not for a specific shared mobility service) ........................................ 123
4.3.2. Criteria related to traveler choices that are common across stakeholders and
shared mobility services ............................................................................... 127
4.3.3. Summary of the main-criteria and sub-criteria to be considered . 128
Experimental Activities ................................................................................. 129
5.1 Study area ............................................................................................ 129
5.1.1. Shared mobility services in Italy .................................................. 129
5.1.2 Description of the study area and shared mobility services in Turin132
5.2 Questionnaires design .......................................................................... 135

VII
5.2.1 Surveys associated with stakeholders of car-sharing, bike-sharing, and scooter-
sharing services (surveys 1 to 6) ................................................................... 141
5.2.2 Surveys associated with stakeholders of shared mobility service services (as a
whole, not for a specific shared mobility service) (surveys 7 to 9) .............. 142
5.3 Data collection activities ..................................................................... 144
5.4 Collected data ...................................................................................... 145
5.4.1 Socio-demographic characteristics of users and non-users .......... 146
5.4.2. Routines and daily travel views of users and non-users .............. 147
5.4.3 Selected data (responses to the BWM-related questions) in this study 152
5.4.4. Socio-demographic characteristics of selected users and non-users of each of
the shared mobility services ......................................................................... 155
5.4.5. Views of whole operators and members of the government regarding some of
the travel routines of users of each of the shared transportation services .... 155
Results ............................................................................................................ 156
6.1 Results of the Analysis for Each Shared Mobility Service (Separately)156
6.1.1 Car-sharing services ..................................................................... 157
6.1.2 Bike-sharing services ................................................................... 178
6.1.3 Scooter-sharing services .............................................................. 197
6.1.4 Comparing the relative importance of different criteria among the three types
of shared mobility services ........................................................................... 215
6.2 Results of the Analysis for Shared Mobility Services (as a whole, not for a specific
shared mobility service) ................................................................................... 225
6.2.1 Group weight of government members (shared mobility services as a whole,
not for a specific shared mobility service) .................................................... 226
6.2.2 Group weight of operators of shared mobility services (as a whole, not for a
specific shared mobility service) .................................................................. 228
6.2.3 Group weight of users of shared mobility services (as a whole, not for a
specific shared mobility service) .................................................................. 230
6.2.4 Group weight of non-users of shared mobility services (as a whole, not for a
specific shared mobility service) .................................................................. 232
6.2.5 Similarities and differences between the four types of shared mobility
stakeholders (as a whole, not for a specific shared mobility service) .......... 233
6.2.6 Perception analysis ........................................................................ 235
6.2.7 Sensitivity analysis and scenarios ................................................. 237
Conclusions .................................................................................................... 244

VIII
References ...................................................................................................... 253
Appendix 1: Details on the methodology of the review of the socio-demographic factors
for car-sharing and previous reviews in the same area .................................. 283
Appendix 2: Survey questionnaires ............................................................... 286
A2.1 Questionnaires for users and non-users of each shared mobility service
(surveys 1 to 3) ............................................................................................. 287
A2.2 Questionnaires for government members and operators of each shared mobility
service (surveys 4 to 6) ................................................................................. 305
A2.3 Questionnaires for users and non-users of shared mobility services (as a whole,
not for a specific shared mobility service) (survey 7) .................................. 309
A2.4 Questionnaire for government members about shared mobility services (as a
whole, not for a specific shared mobility service) (surveys 8) ..................... 314
A2.5 Questionnaire for operators of shared mobility services (as a whole, not for a
specific shared mobility service) (survey 9) ................................................. 316
Appendix 3: Codebook .................................................................................. 318
A3.1 The codebook for users and non-users of {car, bike, scooter}-sharing (general
codebook) (surveys 1 to 3) ............................................................................... 318
A3.2 The codebook for government members and operators {car, bike, scooter}-
sharing (general codebook) (surveys 4 to 6) .................................................... 354
A3.3. The codebook for users and non-users of shared mobility services (as a whole)
(survey 7).......................................................................................................... 363
A3.4 The codebook for government members about shared mobility services (as a
whole) (survey 8) ............................................................................................. 376
A3.5 The codebook for operators of shared mobility services (as a whole) (survey 9)
.......................................................................................................................... 380
A3.6 Job positions of government members and operators (surveys 4, 5, 6, 8, and 9)
.......................................................................................................................... 384
Appendix 4: Descriptive statistics of the data set .......................................... 387
A4.1 Socio-demographic characteristics of users and non-users of each of the shared
mobility services .............................................................................................. 387
A4.2 Routines and daily travel views of users and non-users of each of the shared
mobility services .............................................................................................. 391
A4.3 Socio-demographic characteristics of selected users and non-users of each of the
shared mobility services ................................................................................... 407
A4.4 Perspectives of whole operators and members of the government regarding some
of the travel routines of users of each of the shared transportation services .... 416

ix
List of Figures
Figure 1: Structure of the study. .......................................................................................... 6
Figure 2: Various steps of the MAMCA method (Macharis et al., 2010). ........................ 89
Figure 3: RI for different values n (Saaty, 1980). ............................................................. 96
Figure 4: PROMETHEE methodology (Kolios et al., 2016). ........................................... 98
Figure 5: Set of criteria from 1 to n. ................................................................................ 101
Figure 6: Choosing the criteria of the best and the worst. ............................................... 102
Figure 7: The preference of the best criterion over other criteria. .................................. 102
Figure 8: The preference of each criterion over the worst criterion (Rezaei, 2015). ...... 103
Figure 9: The concepts of  , and . ............................................................ 104
Figure 10: The probabilistic graphical model of the Bayesian BWM (Mohammadi and
Rezaei, 2020). ........................................................................................................................ 111
Figure 11: Important stakeholders of shared mobility services. ..................................... 121
Figure 12: Relationship between the stakeholders of shared mobility services. ............. 121
Figure 13: Purpose of analysis of shared mobility services (as a whole, not for a specific
shared mobility service) and an analysis of each shared mobility service (separately). ........ 123
Figure 14: Number and percentage of available services of each shared mobility system
in Italy in 2020 (Ciuffini et al., 2021). ................................................................................... 130
Figure 15: Number and percentage of available vehicles of each shared mobility system
in Italy in 2020 (Ciuffini et al., 2021). ................................................................................... 130
Figure 16: Number and percentage of rentals of each shared mobility system in Italy in
2020 (Ciuffini et al., 2021). ................................................................................................... 131
Figure 17: Map of the district of Turin. .......................................................................... 133
Figure 18: Map of the Traffic Analysis Zones outside the municipality of Turin (Agenzia
per la Mobilità Metropolitana e Regionale, 2015). ................................................................ 134
Figure 19: Stakeholders and the survey associated with each shared mobility service to
which they responded. ........................................................................................................... 137
Figure 20: Screenshot of the survey with BWM-related questions (question set A in
survey 7)................................................................................................................................. 139
Figure 21: Screenshot of the survey with routines and daily travel views questions
(question set B in surveys 1 to 3). .......................................................................................... 139
Figure 22: Screenshot of the survey with socio-demographic characteristics questions
(question set C in surveys 1,2 and 3). .................................................................................... 140

x
Figure 23: Screenshot of the survey with questions about some characteristics that might
induce people to use (or use more) (question set D in surveys 4, 5, and 6). .......................... 140
Figure 24: Screenshot of the survey with questions about some characteristics affecting
the use of shared mobility services (question set E in survey 7). .......................................... 141
Figure 25: A sample of a 7-point semantic scale question (question set B in survey 7). 144
Figure 26: Screenshot from the original online survey (first BWM question (question
B1)) (question set A in survey 1). .......................................................................................... 145
Figure 27: Percentage (as well as the absolute number) of users of each shared mobility
service (question set C in surveys 1 to 3 respondents) living in Turin and outside Turin. .... 147
Figure 28: Percentage (as well as the absolute number) of non-users of each shared
mobility service (question set C in surveys 1 to 3 respondents) living in Turin and outside
Turin. ...................................................................................................................................... 147
Figure 29: The percentage (as well as the absolute number) of users of each shared
mobility service who use and do not use their private car on a daily basis (question set B in
surveys 1 to 3 respondents). ................................................................................................... 148
Figure 30: The percentage (as well as the absolute number) of non-users of each shared
mobility service who use and do not use their private car on a daily basis (question set B in
surveys 1 to 3 respondents). ................................................................................................... 149
Figure 31: Credal ranking of main-criteria from government members’ view for car-
sharing services. ..................................................................................................................... 158
Figure 32: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (car-sharing services). .............................................................. 160
Figure 33: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (car-sharing services). .............................................................. 161
Figure 34: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (car-sharing services). .............................................................. 162
Figure 35: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for car-sharing choice).
................................................................................................................................................ 162
Figure 36: Credal ranking of main-criteria from operators’ view for car-sharing services.
................................................................................................................................................ 163
Figure 37: Credal ranking of sub-criteria belonging to the main-criterion C3 from
operators’ view for car-sharing services. ............................................................................... 164
Figure 38: Credal ranking of sub-criteria belonging to the main-criterion C2 from
operators’ view for car-sharing services. ............................................................................... 165
Figure 39: Credal ranking of sub-criteria belonging to the main-criterion C1 from
operators’ view for car-sharing services. ............................................................................... 166

xi
Figure 40: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of car-sharing operators). ................................ 166
Figure 41: Credal ranking of main-criteria from users’ view for car-sharing services. .. 167
Figure 42: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for car-sharing services. ................................................................................................ 168
Figure 43: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for car-sharing services. ................................................................................................ 169
Figure 44: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of car-sharing services................................................................................................... 169
Figure 45: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of car-sharing). ............................................ 170
Figure 46: Credal ranking of main-criteria from non-users’ view for car-sharing services.
................................................................................................................................................ 171
Figure 47: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-
users’ view of car-sharing services. ....................................................................................... 172
Figure 48: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-
users’ view of car-sharing services. ....................................................................................... 172
Figure 49: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-
users’ view of car-sharing services. ....................................................................................... 173
Figure 50: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of car-sharing). ........................... 173
Figure 51: Importance of main-criteria based on different types of stakeholders. ......... 176
Figure 52: Importance of sub-criteria based on different types of stakeholders. ............ 177
Figure 53: Credal ranking of main-criteria from government members’ view for bike-
sharing services. ..................................................................................................................... 178
Figure 54: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (bike-sharing services). ............................................................ 180
Figure 55: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (bike-sharing services). ............................................................ 180
Figure 56: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (bike-sharing services). ............................................................ 181
Figure 57: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for bike-sharing
choice). ................................................................................................................................... 182
Figure 58: Credal ranking of main-criteria from operators’ view for bike-sharing services.
................................................................................................................................................ 182

xii
Figure 59: Credal ranking of sub-criteria belonging to the main-criterion C3 from
operators’ view for bike-sharing services. ............................................................................. 184
Figure 60: Credal ranking of sub-criteria belonging to the main-criterion C2 from
operators’ view for bike-sharing services. ............................................................................. 185
Figure 61: Credal ranking of sub-criteria belonging to the main-criterion C1 from
operators’ view for bike-sharing services. ............................................................................. 185
Figure 62: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of bike-sharing operators). .............................. 186
Figure 63: Credal ranking of main-criteria from users’ view for bike-sharing services. 187
Figure 64: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for bike-sharing services. .............................................................................................. 188
Figure 65: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for bike-sharing services. .............................................................................................. 188
Figure 66: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of bike-sharing services. ................................................................................................ 189
Figure 67: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of bike-sharing). .......................................... 189
Figure 68: Credal ranking of main-criteria from non-users’ view for bike-sharing
services. .................................................................................................................................. 190
Figure 69: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-
users’ view of bike-sharing services. ..................................................................................... 191
Figure 70: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-
users’ view of bike-sharing services. ..................................................................................... 192
Figure 71: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-
users’ view of bike-sharing services. ..................................................................................... 192
Figure 72: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of bike-sharing). ......................... 193
Figure 73: Importance of main-criteria based on different types of stakeholders. ......... 195
Figure 74: Importance of sub-criteria based on different types of stakeholders. ............ 196
Figure 75: Credal ranking of main-criteria from government members’ view for scooter-
sharing services. ..................................................................................................................... 197
Figure 76: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (scooter-sharing services). ....................................................... 199
Figure 77: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (scooter-sharing services). ....................................................... 199

xiii
Figure 78: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (scooter-sharing services). ....................................................... 200
Figure 79: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for scooter-sharing
choice). ................................................................................................................................... 200
Figure 80: Credal ranking of sub-criteria belonging to the main-criterion C3 from
operators’ view for scooter-sharing services. ........................................................................ 202
Figure 81: Credal ranking of sub-criteria belonging to the main-criterion C1 from
operators’ view for scooter-sharing services. ........................................................................ 202
Figure 82: Credal ranking of sub-criteria belonging to the main-criterion C2 from
operators’ view for scooter-sharing services. ........................................................................ 203
Figure 83: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of scooter-sharing operators). ......................... 204
Figure 84: Credal ranking of main-criteria from users’ view for scooter-sharing services.
................................................................................................................................................ 205
Figure 85: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for scooter-sharing services........................................................................................... 206
Figure 86: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of scooter-sharing services. ........................................................................................... 206
Figure 87: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for scooter-sharing services........................................................................................... 207
Figure 88: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of scooter-sharing). ..................................... 207
Figure 89: Credal ranking of main-criteria from non-users’ view for scooter-sharing
services. .................................................................................................................................. 208
Figure 90: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-
users’ view of scooter-sharing services. ................................................................................ 209
Figure 91: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-
users’ view of scooter-sharing services. ................................................................................ 210
Figure 92: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-
users’ view of scooter-sharing services. ................................................................................ 210
Figure 93: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of scooter-sharing). .................... 211
Figure 94: Importance of main-criteria based on different types of stakeholders. ......... 213
Figure 95: Importance of sub-criteria based on different types of stakeholders. ............ 214
Figure 96: Importance of main-criteria based on different shared mobility services from
the government members' views. ........................................................................................... 216

xiv
Figure 97: Importance of sub-criteria based on different shared mobility services from the
government members' views. ................................................................................................. 217
Figure 98: Importance of main-criteria based on different shared mobility services from
the operators' views. ............................................................................................................... 218
Figure 99: Importance of sub-criteria based on different shared mobility services from the
operators' views. ..................................................................................................................... 219
Figure 100: Importance of main-criteria based on different shared mobility services from
the users' views. ..................................................................................................................... 220
Figure 101: Importance of sub-criteria based on different shared mobility services from
the users' views. ..................................................................................................................... 221
Figure 102: Importance of main-criteria based on different shared mobility services from
the non-users' views. .............................................................................................................. 222
Figure 103: Importance of sub-criteria based on different shared mobility services from
the non-users' views. .............................................................................................................. 223
Figure 104: Credal ranking of criteria from government members’ view for shared
mobility services. ................................................................................................................... 227
Figure 105: The weight of the least important criterion and the three most important
criteria (from the perspective of government members for shared mobility choice). ........... 228
Figure 106: Credal ranking of criteria from operators’ view for shared mobility services.
................................................................................................................................................ 229
Figure 107: The weight of the least important criterion and the three most important
criteria (from the perspective of shared mobility operators). ................................................ 230
Figure 108: Credal ranking of criteria from users’ view for shared mobility services. .. 231
Figure 109: The weight of the least important criterion and the three most important
criteria (from users' perspective of shared mobility services). .............................................. 232
Figure 110: Credal ranking of criteria from non-users’ view for shared mobility services.
................................................................................................................................................ 233
Figure 111: The weight of the least important criterion and the three most important
criteria (from the perspective of non-users of shared mobility services). .............................. 233
Figure 112: Importance of criteria based on users and non-users stakeholders. ............. 235

XV
List of Tables
Table 1: The positive relationship of being a man or a woman with car membership,
usage, or attitude. ..................................................................................................................... 16
Table 2: The positive correlation between young age groups and car-sharing membership,
usage, or attitudes..................................................................................................................... 18
Table 3: The positive correlation between well-educated background and car-sharing
membership, usage, or attitudes. .............................................................................................. 21
Table 4: The positive relationship of occupation and economic status groups on car-
sharing membership, usage, or attitudes. ................................................................................. 23
Table 5: The positive correlation between small household size and car-sharing
membership, usage, or attitudes. .............................................................................................. 25
Table 6: The positive correlation between being single and car-sharing membership,
usage, or attitudes..................................................................................................................... 26
Table 7: Effect of the presence of children on car-sharing membership, usage, or
attitudes. ................................................................................................................................... 27
Table 8: Positive correlation between low vehicle ownership and car-sharing
membership, usage, or attitudes. .............................................................................................. 27
Table 9: The positive correlation between shorter travel time and car-sharing usage. ..... 32
Table 10: The positive relationship between different trip distance ranges and car-sharing
usage. ....................................................................................................................................... 33
Table 11: The positive correlation between weekend traveling, off-peak hours, or in the
morning and car-sharing usage. ............................................................................................... 33
Table 12: Impact of different trip purpose groups to use car-sharing. .............................. 34
Table 13: The positive correlation between the low travel cost and car-sharing use. ....... 35
Table 14: Impact of different land-use patterns to use car-sharing. .................................. 35
Table 15: Impact of different accessibility conditions to use car-sharing. ........................ 36
Table 16: The positive correlation between larger and older stations and car-sharing
usage. ....................................................................................................................................... 37
Table 17: The positive correlation between user satisfaction and car-sharing usage. ..... 38
Table 18: The positive correlation between the high level of environmental concerns and
the importance of social impacts and car use. .......................................................................... 38
Table 19: The positive correlation between previous experience and car-sharing usage. 39
Table 20: The negative correlation between private car symbol status and car-sharing
usage. ....................................................................................................................................... 39

XVI
Table 21: The positive correlation between sense of ownership and car-sharing usage. .. 39
Table 22: Interactions matrix between factors on the use of car-sharing. ......................... 41
Table 23: The effect of different factors on bicycle use. .................................................. 54
Table 24: Factors affecting bike-sharing choice. .............................................................. 67
Table 25: Influence of factors on the use of e- scooter-sharing. ....................................... 80
Table 26: Criteria and sub-criteria influencing the use of each shared mobility system. . 83
Table 27: RI for different values n (Saaty, 1980).............................................................. 95
Table 28: Some of the studies that applied BWM........................................................... 100
Table 29:  (max ) according to the  (Rezaei, 2015). .......................................... 105
Table 30:  thresholds based on the number of criteria and  (Liang et al., 2020).
................................................................................................................................................ 108
Table 31: Description for each CL range for a threshold value of 50. ............................ 114
Table 32: Evaluation of MCDM methods. ...................................................................... 116
Table 33: Symbolize each criterion associated with users and non-users. ...................... 125
Table 34: Symbolize each criterion associated with government members. .................. 126
Table 35: Symbolize each criterion associated with operators. ...................................... 126
Table 36: The three main-criteria and twelve sub-criteria that are common across
stakeholders and shared mobility services. ............................................................................ 128
Table 37: The ratio of the subscribers of each shared mobility service to the population of
the province and city (Ciuffini et al., 2021). .......................................................................... 131
Table 38: The number of survey responses requested (to SWG) and received from the
stakeholders of each shared mobility service (surveys 1 to 9). .............................................. 145
Table 39: Number of responses that passed quality checks from each stakeholder for the
main-criteria and each sub-criteria set for the car-sharing, out of the total number of
responses shown in the last column (question set A in surveys 1 and 4). ............................. 153
Table 40: The number of used responses from each stakeholder for the main-criteria and
each sub-criteria set for the bike-sharing (question set A in surveys 2 and 5). ..................... 154
Table 41: The number of used responses from each stakeholder for the main-criteria and
each sub-criteria set for the scooter-sharing (question set A in surveys 3 and 6). ................. 154
Table 42: The number of used responses from each stakeholder of the shared mobility
services (as a whole) (question set A in surveys 7 to 9). ....................................................... 154
Table 43: Government members’ group weights of the main-criteria for car-sharing
services. .................................................................................................................................. 157
Table 44: The optimal groups’ weights of government members in each sub-criterion for
car-sharing services. ............................................................................................................... 159

XVII
Table 45: Operators’ group weights of the main-criteria for car-sharing services. ........ 162
Table 46: The optimal groups’ weights of operators in each sub-criterion for car-sharing
services. .................................................................................................................................. 163
Table 47: Users’ group weights of the main-criteria for car-sharing services. ............... 166
Table 48: The optimal groups’ weights of users in each sub-criterion for car-sharing
services. .................................................................................................................................. 167
Table 49: Non-users’ group weights of the main-criteria for car-sharing services. ........ 170
Table 50: The optimal groups’ weights of non-users in each sub-criterion for car-sharing
services. .................................................................................................................................. 171
Table 51: Ranking of the main-criteria and sub-criteria corresponding to car-sharing
stakeholders............................................................................................................................ 174
Table 52: Government members’ group weights of the main-criteria for bike-sharing
services. .................................................................................................................................. 178
Table 53: The optimal groups’ weights of government members in each sub-criterion for
bike-sharing services. ............................................................................................................. 179
Table 54: Operators’ group weights of the main-criteria for bike-sharing services. ...... 182
Table 55: The optimal groups’ weights of operators in each sub-criterion for bike-sharing
services. .................................................................................................................................. 183
Table 56: Users’ group weights of the main-criteria for bike-sharing services. ............. 186
Table 57: The optimal groups’ weights of users in each sub-criterion for bike-sharing
services. .................................................................................................................................. 187
Table 58: Non-users’ group weights of the main-criteria for bike-sharing services. ...... 190
Table 59: The optimal groups’ weights of non-users in each sub-criterion for bike-sharing
services. .................................................................................................................................. 190
Table 60: Ranking of the main-criteria and sub-criteria corresponding to bike-sharing
stakeholders............................................................................................................................ 193
Table 61: Government members’ group weights of the main-criteria for scooter-sharing
services. .................................................................................................................................. 197
Table 62: The optimal groups’ weights of government members in each sub-criterion for
scooter-sharing services. ........................................................................................................ 198
Table 63: Operators’ group weights of the main-criteria for scooter-sharing services. .. 200
Table 64: The optimal groups’ weights of operators in each sub-criterion for scooter-
sharing services. ..................................................................................................................... 201
Table 65: Users’ group weights of the main-criteria for scooter-sharing services. ........ 204
Table 66: The optimal groups’ weights of users in each sub-criterion for scooter-sharing
services. .................................................................................................................................. 205

XVIII
Table 67: Non-users’ group weights of the main-criteria for scooter-sharing services. . 207
Table 68: The optimal groups’ weights of non-users in each sub-criterion for scooter-
sharing services. ..................................................................................................................... 208
Table 69: Ranking of the main-criteria and sub-criteria corresponding to scooter-sharing
stakeholders............................................................................................................................ 211
Table 70: Government members’ group weights of criteria for shared mobility services.
................................................................................................................................................ 226
Table 71: Operators’ group weights of the criteria for shared mobility services. ........... 228
Table 72: Users’ group weights of the criteria for shared mobility services. ................. 230
Table 73: Non-users’ group weights of the criteria for shared mobility services. .......... 232
Table 74: Stakeholders, criteria, and related weights. ..................................................... 234
Table 75: Scores  obtained from users and non-users of each shared mobility service.
................................................................................................................................................ 236
Table 76: Perception of the value of each shared mobility service for users. ................. 237
Table 77: Perception of the value of each shared mobility service for non-users........... 237
Table 78: New indicator values for users' perception of the overall value of each shared
mobility service. ..................................................................................................................... 238
Table 79: New indicator values for non-users' perception of the overall value of each
shared mobility service. ......................................................................................................... 239
Table 80: New perception of the overall value of each shared mobility service analysis
results for users. ..................................................................................................................... 239
Table 81: New perception of the overall value of each shared mobility service analysis
results for non-users. .............................................................................................................. 239
Table 82: Current situation and possible scenarios for the users’ perception of the overall
value of each shared mobility service and the corresponding scenarios ranks (as a whole, not
for a specific shared mobility service). .................................................................................. 240
Table 83: Current situation and possible scenarios for the non-users’ perception of the
overall value of each shared mobility service and the corresponding scenarios ranks (as a
whole, not for a specific shared mobility service). ................................................................ 241
Table 84: Suggestions for government members and operators to pay more attention (+)
(because they underestimate) or less attention (-) (because they overestimate) to the
importance of the main-criteria. ............................................................................................. 249
Table 85: Suggestions for government members and operators to pay more attention (+)
(because they underestimate) or less attention (-) (because they overestimate) to the
importance of sub-criteria. ..................................................................................................... 249

1
Chapter 1
Introduction
Recent decades have seen changes in the way urban transportation is viewed. Initially, the
rising use of private transportation in industrialized countries provided greater access.
However, it has led to negative externalities such as pollution and excessive energy and time
consumption in the long run because of traffic congestion. Mainly this is more likely to occur
in urban areas where demand is concentrated during peak hours (Jorge and Correia, 2013).
Furthermore, car ownership costs such as fuel, parking, and the cost of car insurance are rising
(Mitchell et al., 2010). Public transportation could be a proper alternative, but it has several
drawbacks. For example, public transport coverage does not provide door-to-door service, even
in European cities with a significant public transport network. Also, public transport service
lacks personalization and a flexible schedule (Jorge and Correia, 2013).
International concerns over climate change and global motorization have heightened
interest in sustainable transportation strategies. These include integrated land use and
transportation plans, vehicle technologies, clean fuels, and transportation demand management
(Shaheen and Lipman, 2007). Urban transportation systems face challenges such as
accelerating population growth, urban sprawl, congestion, and overcrowded public
transportation services. The level of service provided by conventional modes of transport is
affected by these problems and inevitably intensifies dependence on a private vehicle. Under
these circumstances, the transportation market is fundamentally changing. It provides new
opportunities for more flexible, efficient, and responsive solutions, such as introducing shared
mobility modes of car-sharing systems scattered around a city (Calderón and Miller, 2020).
The term 'shared mobility' contains car-sharing modes, private vehicle sharing (fractional
ownership, peer-to-peer car-sharing), scooter-sharing (in Italy, it is called "Sharing di
Monopattini Elettrici”), traditional ridesharing, bicycle-sharing, transport network companies
(ride-sourcing), and Electronic hailing (taxis). In addition, it can encompass flexible transit
services, consisting of micro transits that complement rail and fixed-track bus systems
(Shaheen and Chan, 2016).
The car-sharing system consists of a small and medium fleet of cars available at several
stations that can be used by a relatively large group of members (Shaheen et al., 1999). The
car-sharing system is a mode of transportation that combines the freedom of a private car and

2
the affordable cost of traditional public transit (Barth and Shaheen, 2002; Martin and Shaheen,
2011a; Habib et al., 2012; Morency et al., 2012; Uteng et al., 2019; Ceccato and Diana, 2021).
Furthermore, the car-sharing system can offer privacy and flexibility as a private car and also
does not have the disadvantages of public transportation (Barth and Shaheen, 2002; Zhou and
Kockelman, 2011; Clewlow, 2016) without directly incurring all costs (Cooper et al., 2000;
Huwer, 2004; Shaheen et al., 2006; Martin and Shaheen, 2011a; Costain et al., 2012; De
Lorimier and El-Geneidy, 2013; Shaheen and Cohen, 2013; Efthymiou et al., 2013; De Luca
and Di Pace, 2015; Shaheen and Chan, 2016; Efthymiou and Antoniou, 2016; Yoon et al.,
2017; Wang et al., 2017; Kim et al., 2017a; Hua et al., 2019; Jones and Leibowicz, 2019) and
restrictions (Coll et al., 2014), it can bridge the gap between private car and public transport
(Morency et al., 2007; Efthymiou et al., 2013; Kaspi et al., 2014).
Car-sharing systems can have positive effects on the environment. Generally, the car-
sharing system has positively affected urban mobility since each car is used more efficiently
than private vehicles (Litman, 2000; Schuster et al., 2005). The utilization rate of shared
vehicles is more than single-user private vehicles due to spending less time in the parking lot
and more time on the road, leading to less sunk costs. In addition, less land is needed for car
parking in the medium and long term (Mitchell et al., 2010). Hence, the car-sharing program
is an opportunity to develop sustainable urban development (Costain et al., 2012; Jorge,
Barnhart, and de Almeida Correia, 2015) without the obligation of passengers to relinquish the
benefits of using the private car (Huwer, 2004). It is important to note that car-sharing service
does not eliminate car use, but it does make individuals aware of how to use the car properly
(Huwer, 2004; De Lorimier and El-Geneidy, 2013; Coll et al., 2014; Morency et al., 2015). In
developed countries, many young people postpone obtaining a driver's license (Mounce and
Nelson, 2019). Furthermore, it indicates that the importance of having a car is gradually
diminishing (Schmöller et al., 2015). These reasons shift from car ownership to "car as
demand" (Firnkorn and Müller, 2012; Kent and Dowling, 2016; Mounce and Nelson, 2019).
Global mobility challenges are omnipresent. Urban centers worldwide face challenges
with a lack of space, congestion, and high emissions levels (Gössling, 2020; Maiti et al., 2020).
Micro-mobility is a promising urban mobility solution (Feng et al., 2020). The term micro-
mobility refers to incorporating a short trip with a small vehicle. It is called micro-mobility
when transport mobility is restricted to only a limited range of travel for light vehicles
(Elhenawy et al., 2020). Shared micro-mobility is those services providing short-term electric
rental vehicles to the general public for a fee (McKenzie, 2019). Vehicles of light categories
such as motorbikes, e-bikes, electric scooters (e-scooters), bikes, shared bikes, and some riding
devices such as skateboards are considered micro-mobility vehicles (Tuncer et al., 2020).
A Bike-Sharing System (BSS) is a new flexible form of investment in contiguous
bicycle infrastructure that has been theorized to encourage more bike trips (Buck and Buehler,
2012; Gleason and Miskimins, 2012). The BSS is defined as the shared utilization of a bike
system in which users have access to the fleet of bikes offered in public space (Büttner and
Petersen, 2011). The recent proliferation of the bike-share scheme (BSS), recognized as a
public bike use program, is one of the sustainable transportation alternatives that assist in
alleviating the abovementioned concerns (Bauman et al., 2017). For instance, since the BSS is

3
an eco-friendly and emission-free transport mode, it could provide a low-carbon solution for
the “last mile” problem (DeMaio, 2009; Zhang et al., 2015). The last mile is traveling the short
distance between transit stations, the workplace or home, and public transport that is too far for
walking (Zhang et al., 2015).
Furthermore, the other benefit is cost savings from the modal shift. It enables individuals
to use bikes at an affordable cost, with fewer responsibilities compared to bike ownership.
Generally, the bike-sharing mode is known as an affordable means of transportation (Hyland
et al., 2018). In addition, bike-sharing as a daily mode of transportation can help alleviate fuel
costs, curb traffic congestion, and increase health benefits and environmental awareness
(Shahin et al., 2010, Fishman et al., 2013; Li et al., 2019). BSSs triggered 25,240 tons of carbon
dioxide and 64 tons of nitrogen oxide gas emissions to fall and reduced 8358 tons of gasoline
consumption, conferring improved air quality in Shanghai, China, in 2016 (Zhang and Mi,
2018). Hence, increasing the deployment and utilization of the BSS can yield notable
greenhouse gas emission reductions and provide obvious environmental benefits (Bajracharya
et al., 2018). Also, implementing a Public Bike-Sharing (PBS) program turned out to be
effective in raising the cycling rate among people living in areas where the BSSs are available
(Fuller et al., 2013; Ricci, 2015; Godavarthy and Taleqani, 2017).
Bike reservation, pick-up, and drop-off processes in this system are self-service. BSS is
commonly concentrated in urban settings, with lower implementation and operational costs
(e.g., in contrast to shuttle services). There are two types of bike-sharing stations. First, there
is a bike-sharing service in which the Bike-Sharing Program (BSP) provides multiple dock
(fixed stations that lock the bicycles and release the bike by computer control) locations that
enable people to pick-up and drop-off bikes at the different docks. Another service is dockless
(flex stations), where people can receive a code on their mobile phone to unlock the bike and
pick-up the bike or drop-off in a public place, so there is no need for docking stations (Shaheen
et al., 2010).
E-scooters are part of micro-mobility, complementing existing transport networks
(Button et al.,2020). Micro-mobility may alleviate some challenges facing today's large cities
and provide a sustainable urban transport path. Shared stand-up e-scooters (electric
kick/standing scooters) are shared micro-mobility. It is important to note that e-scooters should
not be confused with the small electric motorcycles on which motorcyclists sit, as they are
sometimes called e-scooters. Standing e-scooters are similar to children's but are equipped with
small motors (Button et al., 2020). They are available in many cities as short-term rental options
(Hollingsworth et al., 2019). Especially with the change to the Mobility-as-a-Service paradigm,
e-scooter-sharing has become a common means of transportation in cities (Ciociola et al.,
2020).
E-scooters are battery-powered, motorized versions of kick-scooters and have a long and
narrow platform on which users can stand. There is also a vertical pole at the front with
handlebars, throttle, brake controls, and two small in-line wheels at the front and rear (Fang et
al., 2018). E-scooters are small, electric, and single-occupancy vehicles that are part of the
global boom in "urban micro-mobility" (Tuncer and Brown, 2020). The contribution of e-

4
scooters can play an essential role in improving accessibility in less-connected communities
and supporting transportation sustainability (Zou et al., 2020). E-scooters are better than cars
in terms of ecological potential (Kazmaier et al., 2020). However, due to the limited charge of
the scooters' batteries, the distance traveled by e-scooters is limited. Otherwise, if the charge
falls below the use conditions, passengers must leave the shared scooter halfway to the
destination (Zhu et al., 2020). E-scooters are usually dockless, meaning there is no fixed
location, and they are picked-up and dropped-off from arbitrary places in the service area
(Fawcett et al., 2018). With the Internet of Things (IoT), mobile payment, and location-based
services, dockless e-scooter-sharing does not require fixed docking stations for users (He and
Shin, 2020). The user accesses the available scooters through a special service program
downloaded on their mobile device. After finding the available e-scooter, the user scans the
Quick Response (QR) code on the e-scooter, opens it, and starts the trip. After reaching the
destination, the user can park the e-scooter, click the mobile application's end button, and leave
the e-scooter. The travel cost is charged to the credit card linked to the mobile app (McKenzie,
2019). Data connections, location data of GPS units, and mobile apps are utilized to prevent
theft, help users find the e-scooter to rent, and allow companies to collect scooters for service
or charging (Petersen, 2019).
People can move on city streets by e-scooters, addressing mobility problems such as
congestion and the first and last mile (Bai and Jiao, 2020). The e-scooter is a competitive mode
of transportation in last-mile situations (Baek et al., 2021). The advantages of e-scooters could
vary significantly in geographical areas only a few blocks away because of the differential
access to bus routes and transit lines (Smith and Schwieterman, 2018). E-scooters can help
people who live farther away from such stations access them more quickly, thus encouraging
multimodal travel. The dimensions of an e-scooter spatially take up a little more space than a
pedestrian but occupy much less than a cyclist (Tubis et al., 2019). Because of the accumulation
at traffic junctions, e-scooters can offer an easy solution if the destination is not appropriately
connected to the public transportation network or for long distances that seem like a long walk
(Allem and Majmundar, 2019). Also, e-scooter users can benefit from low travel costs due to
by-the-minute e-scooter rental services and healthy competition between micro-mobility
service providers (Maiti et al., 2020).
Since shared mobility systems are a mode of transportation that combines the
advantages of private vehicles and transit services, policy-makers might not know how to treat
these kinds of services well. Furthermore, although many policies promoting shared mobility
use have been proposed, they have less impact on triggering passengers to shift mode from
private vehicles to shared mobility. It might be because the real requirements of passengers
towards transport mode in the shared mobility service are not well understood. Hence, it is
important to figure out what should be improved in shared mobility services. Also, it is
important to understand the existing different views between users and service providers.
Besides, the difference between the perspectives of users and non-users should be determined
to be able to not only attract users to use the service more but also induce non-users to choose
this service as their transport mode. Also, the requirements for transport modes are abstracted
into a set of factors, and the perceived importance is assigned to each factor. Hence, it is

5
necessary to identify the gap between the needs, expectations, and views of different
stakeholders in car-sharing, bike-sharing, and scooter-sharing systems. To do this, five research
questions can be introduced as follows.
1. Do perceptions vary across stakeholders regarding each criterion in each shared
mobility system?
2. Is there a difference in the importance of criteria between bike-sharing, car-sharing,
and scooter-sharing systems?
3. How do different shared mobility service stakeholders score the importance of
different comparison factors associated with each stakeholder?
4. Which shared mobility system is the best suited for implementation according to the
users’ and non-users’ perceptions in the Turin metropolitan area?
5. Once having clarified the relative importance of different criteria, how can such results
be used to improve sustainable transportation systems?
The present thesis work is structured to address the above questions. First, an introduction
briefly explains each shared mobility service, the benefits of using these services, and the five
research questions investigated in this study. Then, a literature review is conducted for each
shared transportation service to determine what factors influence demand. Some factors that
are important from the author's view but have not been well investigated in the literature are
considered in this research. After that, in the methodology section, a suitable method is selected
according to the research questions and the purpose of the research. Then, in the
implementation of the method, the various stages of the performance of the selected method
are explained. Next, the process of data selection and description of the obtained data is done
in the experimental activities section. After this, the results obtained from the methods are
given in the results section. Finally, in the end, a review of the key obtained conclusions,
recommendations to government members and operators, limitations of the study, and
suggestions for future studies are provided. The overall structure of the study is illustrated in
Figure 1.

6
Figure 1: Structure of the study.

7
Chapter 2
Literature Review
According to the research questions and purposes of this study mentioned in Chapter 1, this
section aims to deliver an overview of car-sharing, bike-sharing, and scooter-sharing services
to understand better the important criteria and sub-criteria that can affect each of these shared
mobility services. The reasons for placing the sub-criteria in each criterion are based on the
literature, the author’s knowledge, as well as the similar characteristics of those sub-criteria.
2.1 An overview of car-sharing
The car-sharing system and its benefit are explained in Chapter 1. This section provides an
overview of car-sharing services to understand better the important criteria and sub-criteria that
can affect car-sharing usage. In this regard, explanations about the history, trends,
classification, interaction with other modes of transportation, factors affecting demand,
interaction effects among different factors, and a summary of the description are provided as
follows.
2.1.1 History and trends of car-sharing systems
Technological advances help expand the concepts of a shared economy, a developing
phenomenon that favors the shift from private to service (shared mobility) (Vosooghi et al.,
2017). Technologies such as social networking, location-based services, the Internet, electric
vehicles, access to keyless vehicles, in-car navigation systems, and mobile GPS allow operators
and users to track the location of the car (Kaspi et al., 2014; Shaheen and Chan, 2016) have
played an essential role in the growth of the car-sharing system over time (Morency et al., 2015;
Shaheen et al., 2016; Becker et al., 2017a; Lempert et al., 2019; Standing et al., 2019).
According to Shaheen et al. (1999), the first shared vehicles were mainly created for economic
reasons. These origins can be traced back to 1948, when the Sefage Cooperative launched its
services in Zurich, Switzerland. Elsewhere, a series of "public car" tests were unsuccessful.
Amongst these failures was the Procotip, a car-sharing initiative launched in 1971 in
Montpellier, France. Another case was Witkar, which was settled in Amsterdam in 1973.
However, the experience gained from failures and advances in communication technology
launched several successful programs in the 1980s. These included Mobility Car-sharing in

8
Switzerland and Stattauto in Germany. It was initially anticipated that car-sharing would not
work in the United States because, as Fishman and Wabe (1968) noted, 'American cities have,
with almost no exception, become "motor" cities – adapted to the owner-driver form of
transport.' Hence, car-sharing schemes only appeared under the Mobility Enterprise program
in the 1980s.
Compared to early European users, those living in the United States prefer convenience to
affordable prices, probably owing to inexpensive driving in the United States (Lane, 2005). In
the 1990s, shared vehicles became prevalent in the United States. Several experimental
programs were performed to understand how to run and operate this system. These comprise
Carlink I and II at the Bay Area Rapid Transit station in Dublin-Pleasanton, ZEV.NET at the
University of California, Irvine, and UCR Intellishare at the University of California, Riverside
(Shaheen et al., 2000; Shaheen and Wright, 2001). These programs brought insights into user
behavior in shared vehicles and assessed the feasibility of these systems as a business.
Therefore, in many countries such as Japan, the United States, and Singapore, the natural
progression toward commercializing this concept was predicted (Kek et al., 2006). Although
the first car-sharing partnership was launched in 1948 (Shaheen et al., 1999; Shaheen and
Cohen, 2007; Becker et al., 2017a), the car-sharing system has only expanded in recent years
(Morency et al., 2015; Clewlow, 2016; Lempert et al., 2019), and has become a common mode
of transportation around the world (Shaheen and Cohen, 2007; Costain et al., 2012), even in
Italy (Rotaris et al., 2019). Ceccato's (2020) study in Turin, Italy, concluded that people who
used car-sharing were satisfied with the service and wanted to use it in the future. Especially
on congested streets, car-sharing appeared to be attractive for city travel.
It should be noted that car-sharing systems are different from traditional car rentals because
car-sharing services can provide short-term access (Lagadic et al., 2019). Rates are measured
in minutes or hours (Ciari et al., 2015), not days or weeks (Del Mar Alonso-Almeida, 2019).
In addition, in the car rental process, cars are borrowed on a contract basis and are picked up
from centralized and staffed locations for each rent (Stillwater et al., 2008). Conversely, in
most car-sharing programs, a single contract is set up at the subscription stage (Ceccato, 2020),
and shared cars are reserved and picked up directly by the user (Shaheen et al., 2006; Stillwater
et al., 2008; Shaheen and Chan, 2016; Terrien et al., 2016; Juschten et al., 2017). Shaheen et
al. (2015) define a car-sharing system as short-term access to a car among members who share
a fleet of cars that a third-party organization maintains, operates, and ensures. In a car-sharing
system, users usually have vehicle access by booking them via smartphones or simply picking
up units on the street. Due to real-time vehicle tracking, service providers do not need to do the
matching. Car-sharing operators provide car-sharing services, cars, and maintenance (Huwer,
2004; Shaheen et al., 2006: Kim et al., 2017a; Mounce and Nelson, 2019). The car-sharing
travel cost is calculated based on the trip (Ferrero et al., 2018). It depends on the use of the car
(Efthymiou et al., 2013; Jian et al., 2017), especially the distance and/or travel duration (Huwer,
2004; Stillwater et al., 2008; Efthymiou and Antoniou, 2016; Juschten et al., 2017). Depending
upon the business model, the cost entails insurance, maintenance, parking, membership costs,
fuel, and congestion pricing (Stillwater et al., 2008; Efthymiou et al., 2013; Shaheen and

9
Cohen, 2013; Ciari et al., 2015; Shaheen and Chan, 2016; Efthymiou and Antoniou, 2016; Kim
et al., 2017a; Del Mar Alonso-Almeida, 2019).
2.1.2 Car-sharing classification
Car-sharing is not a univocal concept. Different systems can have widely different travel
demand characteristics, ambits of application, and impacts according to their operational
scheme. Therefore, it is important to consider the different variants implemented in urban areas
worldwide. Car-sharing business models can include four groups: Peer-to-Peer (P2P),
Business-to-Business (B2B), Business-to-Customer (B2C), and Business to Government
(B2G) (Shaheen et al., 2019).
2.1.2.1 Peer-To-Peer (P2P)
Peer-to-Peer (P2P) is a car-sharing system in which car owners can rent their cars to others
when they are not using them (Balac et al., 2015; Li et al., 2018; Shaheen and Cohen, 2020). It
is implemented through a technology platform provided by a facilitating company to bring the
user and owner together and manage the reservation and payment process (Shaheen et al., 2015,
2018; Lagadic et al., 2019). The user can access the car through a specific device or face-to-
face interactions with the owner (Lagadic et al., 2019). However, Calderón and Miller (2020)
believe that this mobility alternative appears to be operationally different from car-sharing
systems. The P2P model is highly flexible at a lower cost than other systems (Shaheen et al.,
2018; Del Mar Alonso-Almeida, 2019). The operator does not bear the cost of maintaining and
purchasing the fleet. Also, car owners do not spend to make their cars attractive and accept
receiving low earnings from sharing their cars because they do not expect to profit (Dill et al.,
2019). Besides, P2P car-sharing systems can overcome the geographical constraints of
traditional car-sharing systems. In particular, to raise revenue, car-sharing operators typically
focus their cars on areas with high potential demand, reducing access to other zones (Dill et
al., 2019). Conversely, the cars in P2P are widespread throughout the city. Moreover, in a P2P
system, the range of cars users can access is usually more remarkable than in other services
(Shaheen et al., 2018).
2.1.2.2 Business-to-Business (B2B)
Business-to-Business (B2B) is another type of car-sharing in which the companies’ employees
are service members. The company or a third-party operator owns and/or manages the fleet
(Lagadic et al., 2019). Thus, this model is characterized by employer-based usage, for example,
for business travels (Fleury et al., 2017) instead of Business-to-Customer (B2C), which has
personal usage (Clark et al., 2015).
2.1.2.3 Buyer-to-Customer (B2C)
In B2C systems, the operators offer public service (Lagadic et al., 2019). This service can be
One-way or Round-trip (Le Vine, Adamou, and Polak, 2014, 2014b; Namazu and Dowlatabadi,
2018; Lempert et al., 2019).
2.1.2.3.1 Round-trip car-sharing system
The Round-trip or Two-way system encompasses home zone-based and Station-based
(Efthymiou and Antoniou, 2016). In Round-trip Station-based services, users pick-up and

10
return the car in the same reserved parking lot (Ferrero et al., 2018; Del Mar Alonso-Almeida,
2019), whereas, in the Round-trip Home Zone-based system, users pick-up and drop-off the
car in the same zone of the city (Firnkorn and Shaheen, 2016). Round-trip car-sharing has been
documented as a strategy to decrease car ownership and mileage in urban areas (Shaheen et al.,
2015). It streamlines the operators' function as demand planning is for each car-sharing station
(Jorge and Correia, 2013). Although daily re-balancing is less critical for service providers in
this system, long-term fleet size decision-makers should pay close attention to users' demands
that car-sharing stations must meet appropriately. One of this service's pros is its reliability
concerning cars and parking space availability (Glotz-Richter, 2016) because car reservations
are made (Le Vine and Polak, 2019).
2.1.2.3.2 One-way car-sharing system
The One-way or point-to-point car-sharing system is station-based and Free-floating (Martin
and Shaheen, 2016; Del Mar Alonso-Almeida, 2019; Lagadic et al., 2019). The One-way
station-based sharing system allows users to return the vehicle to a different car-sharing station
from where it was picked up (Shaheen and Cohen, 2013; Guirao et al., 2018; Ferrero et al.,
2018). In Free-floating programs, users can pick-up and drop-off the car anywhere in a service
area (Becker et al., 2017a, 2017b, 2018; Ferrero et al., 2018). Free-floating is the newest and
most flexible car-sharing system that operates without fixed stations or Round-trip
requirements (Becker et al., 2017a). It has attracted private car owners and public transport
users by providing fast and convenient motorization, especially for short trips (Vosooghi et al.,
2017).
2.1.2.3.3 One-way vs. Round-trip
A car-sharing program has been proposed by some authors that can work with a Round-trip
system under normal conditions and a One-way system for specific locations such as airports,
which can create high demand (Jorge, Barnhart, and de Almeida Correia, 2015). According to
Becker et al. (2017a), Station-based car-sharing is mainly used when individuals require a car.
However, the Free-floating car-sharing system is chosen when it saves time compared to other
alternative modes. The Free-floating car-sharing system is used for a much wider variety of
trips than the Station-based car-sharing system. The Free-floating car-sharing system opens up
car-sharing to One-way travel, i.e., to the airport or commute (Ciari et al., 2014; Le Vine, Lee-
Gosselin, Sivakumar, and Polak, 2015; Becker et al., 2017a). Kaspi et al. (2014) mentioned
that the total excess time users spend in the system can be decreased by 14% to 34% by
incorporating the parking reservation policies in the One-way car-sharing system. Because of
the added flexibility in the One-way car-sharing system, the number of trips generated by One-
way car-sharing systems is three times greater than that of Round-trip systems in Zurich,
Switzerland (Balac et al., 2015). Students also prefer the Free-floating sharing system to the
station in Italy (Rotaris et al., 2019). Moreover, as the One-way system can make commuting
more feasible, it increases service attractiveness (Ciari et al., 2014; Jorge, Molnar, and de
Almeida Correia, 2015).
Accordingly, the Round-trip car-sharing market is relatively small and is mainly used for
leisure, shopping, and sporadic trips (Barth and Shaheen, 2002; Martínez et al., 2017). A survey
performed by Firnkorn and Müller (2011) in Germany confirmed this. The market share of

11
Car2go, a One-way car-sharing company, was approximately 0.37%, which is 25 times more
than that of Round-trip car-sharing. Note, however, that this figure was counted based on the
member subscriptions, not the number of active members. Moreover, in a study by Costain et
al. (2012) in Toronto, Canada, the behavior of a Round-trip car-sharing company was
examined. The results identified that the majority of trip purposes were shopping trips. The
result endorsed the belief that the reasons for travel are limited.
2.1.2.3.4 Re-balancing issue
Notwithstanding that the Free-floating car-sharing service can entice more people, the uneven
demand for this system raises re-balancing challenges for suppliers (Li et al., 2018). Intuitively,
One-way trips inevitably cause some stations to be empty and others to become saturated,
especially during peak hours. In order to overcome the imbalance problem, the dynamic pricing
strategy has been explored as a potential solution (Ciari et al., 2015; Jorge, Molnar, and de
Almeida Correia, 2015). Martínez et al. (2017) proposed a multimodal agent-based
microsimulation for Lisbon, Portugal, and showed that 20% of travel requests were not made
due to a lack of vehicle access. Correspondingly, the re-balancing problems are attributed to
the demand for services during peak hours being three times higher than during off-peak hours.
Hence, a One-way car-sharing system has brought about important operational challenges, such
as parking management and car re-balancing (Shaheen et al., 2015; Brandstätter et al., 2016).
The Free-floating model can include a broader range of trip purposes than the Round-trip
(Jorge and Correia, 2013; Jorge, Barnhart, and de Almeida Correia, 2015). However, due to
the higher flexibility, the re-balancing vehicles for the Free-floating systems are more acute
than their counterparts (Jorge and Correia, 2013; Jorge, Molnar, and de Almeida Correia, 2015;
Terrien et al., 2016). Spatial imbalances are exacerbated because there are no restrictions on
picking-up and dropping-off cars at stations. Spieser et al. (2016) proposed a policy re-
balancing guide for operators, stating that re-balancing-added costs create a trade-off between
financial viability and service quality. Weikl and Bogenberger (2013) mentioned that
consumer-based re-balancing strategies are more prevalent than operator-based approaches in
Free-floating services.
Concerning agent-based microsimulation, Li et al. (2018) presented a supply model of
Free-floating car-sharing by considering the stock of cars in certain places. While it was
assumed that users behave in a First-In, First-Out manner when faced with an under-supply
situation, and individuals park their cars in determined locations. Likewise, in research by Ciari
et al. (2014) in Berlin, Germany, it was determined that the Free-floating service operates well
in complementing Station-based car-sharing systems. Also, a 30% shift from car to Free-
floating car-sharing systems was observed.
Brendel et al. (2018) proposed a decision support system for vehicle relocation, containing
forecasting, relocation, and communication components in the field of re-balancing. The
Econophysics Method was applied to develop a System Energy Relocation Algorithm (SERA)
that first detects cars located in places with low demand and places with a low supply of cars
and afterward comes to the relocation decisions. Also, Wagner et al. (2016) proposed a method

12
based on demand forecasting. A zero-inflated regression describes changes in demand levels
by analyzing the key points of high activity across the city.
2.1.2.4 Business to Government (B2G)
In a B2G model, car-sharing operators provide transportation services to a government agency.
Pricing may include pricing models, such as the per-transaction cost or a fee-for-service
contract. It is important to note that B2G car-sharing services are usually offered by B2C
service operators (Shaheen et al., 2019). Also, since the B2G model is rarely considered in the
literature compared to other business models, it is not reviewed in this study.
2.1.3 Interaction with other modes of transport
Because of the increasing expansion of car-sharing programs, one of the main aspects of
forecasting models is understanding the relationship between car-sharing systems and other
means of transportation (Dias et al., 2017). The ability to demonstrate the nature of this
relationship is significant, given the growing uncertainty of financial resources for
transportation services and the lack of meaningful data presented by private ride-hailing
services. Also, the analysis of complementary and alternative models can contribute to
examining whether the car-sharing system complements or expands existing transport modes
or competes with them for ridership. It can assist policymakers and urban planners in managing
a wide range of mobility alternatives (Welch et al., 2018). Therefore, it is required to gain more
in-depth insight into the relationship between car-sharing and other modes, especially public
transport and private cars.
2.1.3.1 Public transportation and private cars
The Station-based car-sharing system appears to trigger more efficient car usage by gradually
shifting away from private cars to active modes or public transport (Sioui et al., 2013; Becker
et al., 2017a). In contrast, the Free-floating car-sharing system may decline public transport or
active modes in favor of car trips (Firnkorn, 2012; Le Vine, Lee-Gosselin, Sivakumar, and
Polak, 2015; Becker et al., 2017a). This change starts at a high level as the members of the
Free-floating car-sharing system are frequent public transport users. Therefore, this system can
complement public transport (shaheen and Wright, 2001; Huwer, 2004; Shaheen and Martin,
2010; Murphy, 2016; Clewlow, 2016; Becker et al., 2017a; Kim et al., 2017a). Ceccato's (2020)
study in Turin, Italy, confirmed that the car-sharing service could complement public
transportation services. Moreover, the car-sharing system raises public transport usage
(Lempert et al., 2019). Also, the results of some other studies exhibited that there is a
complementary relationship between car-sharing and public transportation systems
(Cervero,2009; Zoepf and Keith, 2016), as they can provide both mobilities for individuals
who do not own a private car (Douma et al.,2008).
From another standpoint, Kortum and Machemehl (2012) noted that high use of transit is
one feature that raises the probability of the city supporting a successful car-sharing scheme.
Car-sharing can respond to the first-mile/last-mile mobility demand (Shaheen and Chan, 2016;
Lagadic et al., 2019). For instance, a Free-floating car-sharing system can be used as a last-
mile connection as part of multi-leg multimodal trips and connect the public transport station

13
and users' final destination (Shaheen and Chan, 2016; Le Vine and Polak, 2019). Also, the
Free-floating car-sharing systems could fill the service gap left by public transport (Becker et
al., 2017a). The car-sharing system can provide access to transit stations in areas where public
transportation is not sufficiently developed, such as rural areas (Cooper et al., 2000; Rotaris
and Danielis, 2018). Also, it can fill a mobility gap for places and times of day that are not
served adequately by transit, such as in off-peak periods or on weekends (Millard-Ball, 2005).
Correspondingly, De Luca and Di Pace (2015) revealed that when public transportation
services are not efficient or guaranteed, the intercity car-sharing plan can complement the
transportation systems.
Acheampong and Siiba (2019) found that dissatisfaction with public transport services lays
the groundwork for car-sharing systems. It means relying on car-sharing systems alone to meet
travel needs without having a comprehensive strategy to provide quality and cost-effective
public transportation services can lead to unsustainable results. Moreover, Efthymiou and
Antoniou (2014) indicated that people who use buses for social trips and those who spend much
time traveling are not satisfied with the car-sharing scheme. Hu et al. (2018) stated that the car-
sharing system appears to have more demand between 1.2 km and 2.4 km from the bus station.
In a different light, Millard-Ball (2005) noted that public transportation could provide easy
access to shared vehicles for passengers away from car-sharing locations. Flexibility in
scheduling and destinations provided by the car-sharing system may be used as a service that
supports the transit by car-sharing users, especially for discretionary trips (Cooper, 2000, Wang
et al., 2017). In a study in Beijing, Yoon et al. (2017) found that individuals who use the buses
or those traveling in a group are more likely to select the Round-trip car-sharing system.
Morsche et al. (2019) found that public transport users were more likely to use flexible public
transport options, while private car drivers were more likely to utilize car-sharing services in
the Netherlands.
Wagner et al. (2015) found that short-distance transport complements car-sharing
activities, while long-distance trains seem to be a substitution. Rotaris et al. (2019) mentioned
that the car-sharing system mainly replaces private cars and, to some extent, public transport.
Ceccato (2020) noted that the car-sharing system might replace trips made by employees and
students on non-working days and weekdays in Turin, Italy. In addition, if the in-vehicle travel
time factor in the car-sharing system is shorter than in public transportation, there may be a
deviation from public transportation to the car-sharing system. Also, it was suggested that to
prevent shifting from public transportation to the car-sharing system; policies should be
considered to maintain short waiting times and low rates, such as raising transportation
frequencies. In addition, public transportation speeds must be enhanced to compete with car-
sharing speeds to reduce potential switches. In a study by Le Vine, Lee-Gosselin, Sivakumar,
and Polak (2015), the P2P car-sharing system was identified as an alternative to public
transportation. In contrast, the Round-trip car-sharing system complemented public
transportation in London.
Furthermore, Ceccato (2020) pointed out a substitution relationship between the car-
sharing system and subway or private cars in Turin, Italy. However, there was no relationship

14
between the car-sharing system and the train, company, or school bus. On the other hand,
waiting time is a factor that affects the shifting from public transportation to car-sharing
systems. If the waiting time for public transportation is more than 3 minutes, favorable switch
rates are expected in Turin, Italy. In addition, potential users are inclined to pay € 0.8 to avoid
4 min wait. In public transportation, a potentially low shift rate was observed for urban travel,
i.e., for short and long distances, especially for less than 10-18 km. In addition, the cost of
public transportation should be lower to avoid switching from public transport to car-sharing
systems. Interestingly, Cervero (2003) mentioned that car-sharing systems are mainly not
attractive in congested areas where transits provide services adequately, such as downtown.
According to Ceccato (2020), car-sharing programs can significantly decrease the number
of car travels in Turin, Italy. Also, decreasing the cost of a car-sharing system could shift from
private cars to a car-sharing program. In addition, the same impact could be enhanced by raising
the cost of driving a private car, decreasing trip time by at least 3 minutes, or declining walking
time to reach a shared car. Also, car-sharing can replace personal car trips less than 14
kilometers, even from outside the city and to destinations within the city. Nevertheless,
potential users are inclined to walk 6 minutes to reach the shared car. Therefore, in order to
increase car-sharing usage, the cost of a car-sharing scheme should not change, but parking
fares should be raised.
2.1.3.2 Walking and bike
Lane (2005) noted that users of car-sharing systems that decreased their car ownership since
joining the car-sharing system drive less (77%), ride bicycles, walk, transit, and use more taxis.
It was also noted that members did not simply replace car trips with trips in car-sharing systems.
Instead, users replaced car travel with a combination of transit, foot, taxi, and to some extent,
bikes. Hence, it was concluded that the car-sharing system could complement walking and
cycling trips, especially for inconvenient activities, by walk and cycling modes, such as night
trips or carrying heavy loads (Cooper et al., 2000). Martínez et al. (2017) noticed that the car-
sharing system is slightly faster on short trips (less than 3 km) than the walking mode. However,
car-sharing has a significantly more significant advantage as the distance traveled increases.
Approximately it is six times faster on long journeys (more than 15 km) than on foot. On the
contrary, due to the attractiveness of the new car-sharing service in San Francisco, people were
more inclined to use the car-sharing system in the first (Cervero, 2003) and second (Cervero
and Tsai, 2004) years instead of walking and cycling. However, members of the car-sharing
system were more likely to use walking and cycling than non-members in the fourth year
(Cervero et al., 2006).
According to Ceccato (2020), the car-sharing system is not appropriate for very short trips,
especially for a distance of fewer than two km and a trip time of fewer than 30 minutes. These
trip types are usually carried out by cycle or on foot. In particular, trips up to 300 meters are
made on foot, while the maximum trip distance by bike is 1.4 kilometers. Moreover, decreasing
the cost of car-sharing trips and walking distance to reach the shared car may induce shifts from
personal cars, cycling, and walking transport modes to the car-sharing system. Nonetheless,

15
cyclists tend to walk up to 9 min, and they may decide to shift if they could decrease this travel
time by at least 5 min compared to the walking time to reach their bicycles.
2.1.3.3 Taxi
Ciari et al. (2015) mentioned a negligible impact of car sharing on taxis in Zurich, Switzerland.
However, Murphy (2016) showed that the car-sharing system is more likely to be substituted
by taxi or car travel than transit travel. Martínez et al. (2017) mentioned that private cars are
more cost-effective than the car-sharing program. Nonetheless, car-sharing systems outdo taxis
in terms of travel costs. Because taxis are subject to night tariffs, but car-sharing systems are
not. A study conducted in five North American cities by Martin and Shaheen (2016) figured
out that members of the car-sharing system reduced taxi use by 42% to 64% after joining the
car-sharing program. Also, Yoon et al. (2017) observed a considerable correlation between taxi
and trip costs, indicating that the car-sharing system can be a competitive alternative mode of
transportation for taxi users, especially when taxi fares are high.
On the other hand, some studies mentioned that the car-sharing system could complement
taxis, which are more suitable for One-way travel and offer an option for individuals who
cannot drive. Also, it can complement the cheaper rental car for long-distance travel (Millard-
Ball, 2005). Efthymiou and Antoniou (2014) found that people who use taxis for social
activities are more likely to use the car-sharing system in Greece.
2.1.4 Factors influencing demand for car-sharing system
As previously mentioned, five factors influencing car-sharing demand can be considered:
socio-demographic characteristics of the traveler, trip-related features, car-sharing
characteristics, built environment and land use characteristics, and attitudinal effects. These are
separately considered in the following subheads. The previous work and methodology of
examining socio-demographic factors for the car-sharing study are described in Appendix 1 as
an example of the review process used in this study.
2.1.4.1 Socio-demographic characteristics influencing the demand for
different car-sharing systems
1
Socio-demographics refer to a combination of socio-demographic factors that define
individuals in a particular group or population. The main socio-demographic factors mentioned
in the literature and considered in this study include gender, age, educational level, occupation
and economic status, household size, marital status, presence of children, and vehicle
ownership status. These are, in fact, the most frequently investigated characteristics in the
reviewed literature. These different social and demographic characteristics can help understand
group members' commonalities (Burghard and Dütschke, 2019). The importance of socio-
demographic factors is that they can be considered key drivers of mobility patterns and travel
modes and can ascertain the diffusion of car-sharing services in the urban population (Prieto et
al., 2017). Generally, a proper understanding of key demographic factors may help increase the
1
Most of the contents of the present section/appendix have been published in Amirnazmiafshar, E., & Diana, M.
(2022). A review of the socio-demographic characteristics affecting the demand for different car-sharing
operational schemes. Transportation Research Interdisciplinary Perspectives, 14, 100616.

16
diffusion of car-sharing services (Millard-Ball, 2005). Focusing on the effect of users' socio-
demographic factors on the choices of different car-sharing operational schemes can help offer
suggestions for the planning and increasing demand for car-sharing operational schemes.
Car-sharing users appear to be a particular group concerning socio-demographics
(Burghard and Dütschke, 2019). People’s features, such as age and gender, can impact member
behavior (Morency et al., 2012). The impact of the main socio-demographic characteristics on
choosing different shared car systems is examined in the following subheads.
Some tables show the impact of the socio-demographic characteristics on the membership
of shared cars, usage, or attitudes in each section. Also, the type of car-sharing services and
any study-specific conditions are shown in the tables to identify the relationship between socio-
economic characteristics and car-sharing demand. Besides, in some tables, the percentage of
members belonging to a particular group or level in each study is specified, as the definitions
in studies are different.
The tables are arranged according to the types of car-sharing services to make them easier
to read. In the row of tables, first, studies on free-floating car-sharing are listed. Then, studies
that have reviewed more than one type of car-sharing service are listed. Finally, studies
examined other car-sharing services, including station-based (service type is not specified),
one-way station-based, P2P, and round-trip station-based are listed.
2.1.4.1.1 Gender
One of the important factors that have been stressed in the previous literature is the gender
factor. Table 1 lists the studies that concluded that either males or females tend to use car-
sharing more consistently.
Table 1: The positive relationship of being a man or a woman with car membership, usage,
or attitude.
Gender
groups
% of members
in this group
Car-sharing
service type
Studied impact
Specific
conditions
Geographic area
References
Male
63.6
Free-floating
Membership
-
Germany
Firnkorn and
Müller, 2012
70.0
Free-floating
Membership
-
Munich and
Berlin, Germany
Kopp et al., 2015
80.0
Free-floating
Membership
Zurich,
Switzerland
Ciari et al., 2015
70.0
Free-floating
Adoption
-
Based,
Switzerland
Becker et al.,
2017a
60.0
Station-based
58.1
One-way
station-based
and free-
floating
Switch from existing
transport mode to car-
sharing
-
Turin, Italy
Ceccato and
Diana, 2021
Unspecified
Station-based
and free-
floating
Switch from existing
transport mode to car-
sharing
-
Ghana, Sub-
Saharan Africa.
Acheampong and
Siiba, 2020
84.6
Round-trip,
free-floating
Membership
-
Berlin, Germany
Kawgan-Kagan,
2015
Unspecified
Station-based
Frequency of use
-
North America
Morency et al.,
2012
74.2
Station-Based
Switch from existing
transport mode to car-
sharing
-
Shanghai, China
Hu et al., 2018
Unspecified
One-way
Station-based
Usage
-
Salerno, Italy
Cartenì et al., 2016

17
Gender
groups
% of members
in this group
Car-sharing
service type
Studied impact
Specific
conditions
Geographic area
References
55.7
Round-trip
Switch from existing
transport mode to car-
sharing
-
Beijing, China
Yoon et al., 2017
About 55.0%
P2P
Membership
-
Portland, USA
Shaheen et al.,
2018
Female
63.0
Free-floating
Membership
-
Montreal, Canada
Wielinski et al.,
2015
51.0
Station-based
57.0
Round-trip
Membership
-
North America
Martin and
Shaheen, 2011a
Unspecified
Round-trip
Membership
New service
San Francisco,
USA
Cervero, 2003
It can be seen that different studies led to different conclusions. It indicates that the gender
dimension is intertwined with other elements that must be considered to clarify how gender
affects car-sharing demand. The first group of studies showed that car-sharing members are
predominantly male (Ciari et al., 2015; Firnkorn and Müller, 2012; Kawgan-Kagan, 2015;
Kopp et al., 2015; Shaheen et al., 2018). Males are more likely to change from their existed
mode of transportation to car-sharing (Acheampong and Siiba, 2020; Cartenì et al., 2016;
Ceccato and Diana, 2021; Hu et al., 2018). Males are more receptive to shared car services,
especially free-floating shared car schemes (Becker et al., 2017a). About 79% of free-floating
service members were male in Turin, Italy (Perboli et al., 2017). In general, males are more
interested in cars, technology, and innovation, of which the car-sharing system is an example
(Kawgan-Kagan, 2020).
Similarly, in Zurich, Switzerland, males accounted for 80% of the free-floating service
members (Ciari et al., 2015). Although males have a higher frequency of use, their trips are
shorter (Habib et al., 2012). Moving from actual behaviors to attitudes, 84% of male users
expressed interest in using car-sharing in a stated preferences survey conducted in Salerno,
Italy. In addition, they raised their utility of switching from personal cars to shared cars (Cartenì
et al., 2016). Morency et al. (2012) indicated that males are more inclined to choose station-
based car-sharing than females in monthly usage. However, although the gender variable was
significant in their study, this parameter’s coefficient was somewhat minor. This reflects the
significant but small impact of gender on station-based car-sharing demand. In Beijing,
although males were more inclined to replace their current mode of transport with round trips,
males and females did not exhibit markedly different behavior on the car-sharing choice for
one-way trips (Yoon et al., 2017).
On the other hand, a handful of papers from North America reported higher membership
rates for females. However, the observed gap was minimal in Martin and Shaheen (2011a).
They focus on round-trip services only, compared to most previously mentioned studies, which
often focused on the correlation between being male and more extensive free-floating services.
In this regard, a study on the willingness to join the round-trip system found no gender
differences (Kim et al., 2017). Cervero (2003) reported a much larger membership rate of
females for a round-trip service in San Francisco, but this could result from the survey being
conducted only one month after the service launch. In addition, this study is significantly older

18
than the average and therefore refers to services whose features differ somewhat from the
contemporary standard practice.
Only one study (Wielinski et al., 2015) reported an over-representation of female members
of the free-floating system in Montreal, which is even more surprising since the gender
distribution in the same city is usually almost the same for different services. Apart from this
exception, about 75% of females chose free-floating services in Berlin, while about 80% of
males did. However, there is a significant gap between females and males for round-trip car-
sharing, while 35% of females chose round-trip car-sharing; this figure was almost 60% for
males. Also, males and females have a similar interest in using e-car sharing. Approximately
80% of females chose Battery Electric Vehicles (BEVs), while 65% chose Internal Combustion
Engine Vehicles (ICEVs) (Kawgan-Kagan, 2015). It indicates that females who chose car-
sharing are more likely to use BEVs instead of ICEVs. However, males chose more ICEVs
than BEVs (Kawgan-Kagan, 2015). Therefore, females seem more attracted to the more
specific BEV systems than the ICEV system (Kawgan-Kagan, 2015; Kim et al., 2015). Del
Mar Alonso-Almeida (2019) offered additional insights into the perceived value role in
increasing female car-sharing demand.
To sum up, males positively correlated with the demand for car-sharing, especially the
free-floating variant, while results are more mixed for round-trip services. However, females
seem keener to choosing e-car-sharing systems. Besides, female car-sharing members in North
American countries appear more inclined to choose car-sharing than female members in
Europe.
2.1.4.1.2 Age
Many studies stated that car-sharing attracted more attention from younger members
(Burkhardt and Millard-Ball, 2006; Ceccato and Diana, 2021; Ceccato, 2020; Firnkorn and
Müller, 2012; Martin and Shaheen, 2011a; Vinayak et al., 2018). Table 2 lists studies stating
that youngsters are more inclined to choose shared cars. Because different articles consider
different definitions of youth, for each study, the age range with the highest percentage of
membership distribution is presented in the first column of Table 2.
Table 2: The positive correlation between young age groups and car-sharing membership,
usage, or attitudes.
Age groups
(brackets or
mean)
% of members
in this group
Car-sharing
service type
Studied impact
Specific
conditions
Geographic
area
References
25-54
77.0
Free-floating
Membership
-
Turin, Italy
Ceccato, 2020
18-34
93.0
Free-floating
Membership
New Service
Turin, Italy
Perboli et al.,
2017
Mean age of
38.7
-
Free-floating
Usage
E-car-sharing
Germany
Burghard and
Dütschke, 2019
Under 35
56.0
Free-floating
Membership
-
Germany
Firnkorn and
Müller, 2012
Under 36
60.0
Free-floating
Membership
-
Austin, USA
Kortum and
Machemehl,
2012
Under 36
50.0
Free-floating
Membership
-
Based,
Switzerland
Becker et al.,
2017a
25-44
73.8
Free-floating
Membership
-

19
Age groups
(brackets or
mean)
% of members
in this group
Car-sharing
service type
Studied impact
Specific
conditions
Geographic
area
References
25-49
71.1
Station-based
Montreal,
Canada
Wielinski et al.,
2015
35-44
25.4
One-way
station-based
and free-
floating
Membership, switch
from existing
transport mode to car-
sharing
-
Turin, Italy
Ceccato and
Diana, 2021
18-24
Unspecified
One-way
station-based
and free-
floating
Membership
-
Seattle. USA
Vinayak et al.,
2018
20-39
About 62.0
Station-based
and free-
floating
Membership
-
Montreal,
Canada
Sioui et al., 2013
The 30s or
40s
Unspecified
Round-trip,
one-way
station-based
Membership
-
North America
Millard-Ball,
2005
18-25
Unspecified
Round-trip,
one-way
station-based,
free-floating,
P2P
Usage
In rural areas
Friuli-Venezia
Giulia, Italy
Rotaris and
Danielis, 2018
The Mid-30s
Unspecified
Round-trip and
one-way
station-based,
B2B
Membership
-
North America
Brook, 2004
31-50
50.0
Station-based
Membership
-
San Francisco
Bay Area, USA
Clewlow, 2016
25–39
55.0
Station-based
Membership
-
Philadelphia,
USA
Lane, 2005
The 20s and
30s
77.9
Station-based
Membership,
willingness to
continue membership
BEV service
Seoul, South
Korea
Kim et al., 2015
25-45
Unspecified
One-way
station-based
Switch from existing
transport mode to car-
sharing
E-car-sharing
Salerno, Italy
Cartenì et al.,
2016
20-35
56.0
One-way
station-based
Interested in car-
sharing
-
Beijing, China
Shaheen and
Martin, 2010
20-40
67.0
Round-trip
Membership
-
North America
Martin et al.,
2010
30-60
55.0
Round-trip
Membership
-
North American
Martin and
Shaheen, 2011a
Mean age of
37.7
-
Round-trip
Membership
-
USA and Canada
Burkhardt and
Millard-Ball,
2006
25-34
55.0
P2P
Membership
-
Portland, USA
Shaheen et al.,
2018
A personal car is no longer a priority for adults, which can be considered a reason to attract
young members to shared cars (Ceccato and Diana, 2021). This shift from car ownership to
“cars as demand” is reinforced by the preference for more sustainable mobility practices
(Ceccato and Diana, 2021; Kortum and Machemehl, 2012). For instance, 67% of car-sharing
members in North America were between 20 and 40 years old (Martin et al., 2010). Also, in
the San Francisco Bay Area, USA, the car-sharing system members are significantly younger
than non-members. About 50% of members are in the age group of 31 to 50 years. However,
this figure is around 37% for non-members (Clewlow, 2016). This may be because the
employment rate among members is higher than among non-members, associated with a lower
average age (Becker et al., 2017a). This is more the case in free-floating car-sharing than in
station-based car-sharing (Becker et al., 2017a; Wielinski et al., 2015). For example, 73.8% of
free-floating members were between 25 and 44 years old in Montreal, Canada. However, the
25 to 49 age group accounted for 71.1% of the members of station-based car-sharing, slightly
less than free-floating. Approximately 93% of the members of free-floating car-sharing were

20
between 18 and 34 years old in Turin, Italy (Perboli et al., 2017). Similarly, half of the free-
floating car-sharing members in Basel, Switzerland, and 56% of members of the system in
Germany were under 36 and 35 years old, respectively (Becker et al., 2017a; Firnkorn and
Müller, 2012).
Car-sharing with Evs has a special added attraction for young couples with no private car.
The same is true for young people who start a family and use car-sharing to complement their
private car trips (Burghard and Dütschke, 2019). In rural areas, similar to urban areas, car-
sharing users are young (Rotaris and Danielis, 2018). In Beijing, China, people encouraged to
use car-sharing belonged to the younger age group of 20 to 35 years (Shaheen and Martin,
2010). Furthermore, 85% of 25–45-year-old people were satisfied using the car-sharing system
in Salerno, Italy (Cartenì et al., 2016). Analogously, some research has shown that members of
shared cars are in their late 20s and mid-30s (Brook, 2004; Lane, 2005) or are 20 to 39 years
old (Kortum and Machemehl, 2012; Sioui et al., 2013) or in their 30s or 40s (Millard-Ball,
2005), or are 25 to 45 years old (Kopp et al., 2015). In Portland, USA, P2P service members
are between 25 and 34 years old. In Switzerland, the effect of age in increasing car-sharing
demand is maximized at age 35 (Juschten et al., 2017). Besides, the older age (55 years or
older) in households without high income negatively affects the willingness to join a car-
sharing program (Dias et al., 2017).
However, Cervero et al. (2007) mentioned that round-trip car-sharing usage increased with
age in San Francisco, USA. Nevertheless, it is significant to stress that this study used the age
factor as a numerical variable. However, in most other studies, age has been used as a class
variable, making it possible to identify potential non-linear relationships. For instance, a study
by Kim et al. (2015) found that 77.9% of e-car-sharing members were within the age group of
the 20s and 30s in Seoul. Interestingly, the probability of switching from private cars to e-car-
sharing among elders is higher than among younger ones. However, this seems to have
happened because the survey is aimed at members of the electric vehicle-sharing program who
have a strong will to change their transportation mode, not the general public. In essence, it can
be indicated that most car-sharing users are young people, typically in their mid-20s to mid-
30s. In addition, free-floating members appear to be slightly younger than station-based
members. Also, it appears that in North America, the age of car-sharing members is a little
older than the age of car-sharing members in other countries.
2.1.4.1.3 Education level
The most prominent feature of car-sharing members is their high education level (Burkhardt
and Millard-Ball, 2006; Becker et al., 2017a; Ceccato, 2020; Firnkorn and Müller, 2012;
Juschten et al., 2017; Kawgan-Kagan, 2015; Shaheen et al., 2018; Shaheen and Martin, 2010).
Table 3 lists the papers that showed that well-educated background raises car-sharing demand.
Different articles have different definitions of well-educated people. For each study, the
educational background of the well-educated people with the highest percentage of
membership distribution is specified in the first column of Table 3.

21
Table 3: The positive correlation between well-educated background and car-sharing
membership, usage, or attitudes.
Education level
% distribution
of the members
Car-sharing
service type
Studied impact
Specific
conditions
Geographic area
References
Master's degree or
PhD
52.9
Free-floating
Membership
-
Turin, Italy
Ceccato, 2020
University degree
or PhD
70.0
Free-floating
Usage
-
Munich and Berlin,
Germany
Kopp et al.,
2015
University or
technical college
46.3
Free-floating
Membership
-
Germany
Firnkorn and
Müller, 2012
Graduate degree
Unspecified
One-way
station-based
and free-
floating
Frequency of use
-
Seattle. USA
Vinayak et al.,
2018
Bachelor’s degree
or higher
Unspecified
One-way
station-based
and free-
floating
Usage
-
Seattle. USA
Dias et al., 2017
University degree
(or equivalent)
75.0
Station-
based
Membership
-
Based, Switzerland
Becker et al.,
2017a
70.0
Free-floating
Graduated from a
university or
technical college
66.7
Round-trip,
free-floating
Membership,
trip frequency
-
Berlin, Germany
Kawgan-
Kagan, 2015
Bachelor’s degree
35.0
Round-trip
One-way
station-based
Membership
-
North America
Millard-Ball,
2005
Postgraduate or
advanced degree
48.0
Upper secondary
education or
higher
71.1
Round-trip,
free-floating,
and P2P
Membership
-
Switzerland
Juschten et al.,
2017
Four-year or
advanced college
graduates
66.7
Round-trip
and one-way
station-
based, B2B
Membership
-
North America
Brook, 2004
Bachelor's degree
or higher
87.0
Station-
based
Membership
-
Portland, USA
Cooper et al.,
2000
Bachelor's degree
or higher
Unspecified
Station-
based
Membership
-
Quebec City,
Canada
Coll et al., 2014
Above high school
diploma
60.0
One-way
station-based
Membership
-
Beijing, China
Shaheen and
Martin, 2010
University
education
Unspecified
Round-trip
Interested in car-
sharing
-
Dublin, Ireland
Carroll et al.,
2017
Bachelor's degree
or higher
84.0
Round-trip
Membership
-
North America
Martin et al.,
2010
Bachelor’s degree
43.0
Round-trip
Membership
-
North America
Martin and
Shaheen, 2011a
Graduate or
professional
degree
41.0
Bachelor’s degree
35.0
Round-trip
Membership
-
USA and Canada
Burkhardt and
Millard-Ball,
2006
Postgraduate or
advanced degree
48.0
Bachelor's degree
or higher
Unspecified
Round-trip
Interested in car-
sharing
-
Shanghai, China
Wang et al.,
2012
Postgraduate
degree
Unspecified
P2P
Adoption
-
Paris, France;
Madrid, Spain;
Tokyo, Japan; and
London, England
Prieto et al.,
2017
Bachelor’s degree
or higher
86.0
P2P
Membership
-
Portland, USA
Shaheen et al.,
2018
A typical figure is that more than sixty-seven percent of members had a bachelor’s or
advanced degree in North America. This rate is remarkably above the average education level
of people living in the neighborhoods where the services are provided (Brook, 2004). Also,
more than 80% of round-trip car-sharing members had a four-year college or advanced degree,
while around 28% of all US citizens had a bachelor’s degree (Martin and Shaheen, 2011a).

22
Similarly, about 87% of station-based car-sharing members had a bachelor’s degree or higher,
while only 31% of Portlanders had a bachelor’s degree (Cooper et al., 2000). This significant
education gap may be because educated people are more adapted to using the internet, such as
booking car-sharing, than others. In addition, these people are usually more prepared to adapt
to a new lifestyle. It is also essential to state that well-educated individuals are associated with
environmental awareness and calculate the car’s actual costs rather than car-sharing (Coll et
al., 2014). Besides, the education level is higher among frequent users of shared transport
(Vinayak et al., 2018). The reason may be that educated decision-makers are more
environmentally friendly and favor a new urban lifestyle. Millard-Ball (2005) suggested that
more than one-third of members in North America have a four-year college degree, and about
half possess a postgraduate or advanced degree. It is noteworthy that an online survey of shared
car members was employed in this study. This survey results primarily represented well-
educated members because they are likelier to use a personal computer. Round-trip car-sharing
members are mostly highly educated (84% have a four-year college or advanced degree) in
North America (Martin et al., 2010).
Beyond car-sharing membership, a high level of education can also increase car-sharing
demand (Coll et al., 2014; Dias et al., 2017; Kopp et al., 2015). It is likely that highly educated
people are more aware of this service and can leverage it through technology (Dias et al., 2017).
This may show that being attracted to car-sharing may be based on a certain level of social
awareness, not strictly an economic decision. Wang et al. (2012) noted that the tendency to use
shared cars is directly related to the level of education. However, this study’s distribution of
academic achievement indicates that this sample had a higher level of education than the
Shanghai population. This may be because the head of the household had filled out the mail
survey, and they probably have the highest education in the household.
Shaheen et al., 2018 found that 86% of the P2P members had bachelor’s degrees or higher.
This may be because P2P car-sharing, like other shared mobilities, operates mainly in urban
areas and larger cities where people with higher education live. However, surprisingly, Prieto
et al. (2017) mentioned that having a higher education level, such as a postgraduate degree,
had no impact on joining P2P car-sharing. This study noted that this is normal because P2P
car-sharing is more compatible with many users. However, it should be noted that the education
factor in this research is insignificant. Most people are looking to choose car-sharing to have a
four-year college degree or higher, especially a postgraduate or advanced degree. Also, it
appears that the education level of round-trip shared car users is less than that of other car-
sharing service users.
2.1.4.1.4 Occupation and economic status
People’s economic and social views can be an important factor influencing their attitudes in
choosing a car-sharing program (Becker et al., 2017a ). Most car-sharing members earn more
than non-members, and most are employed. This may mean that the employee may choose car-
sharing for work-related activities (Ceccato and Diana, 2021; Ceccato, 2020; Clewlow, 2016;
Dias et al., 2017; Juschten et al., 2017; Kawgan-Kagan, 2015; Vinayak et al., 2018; Winter et
al., 2017; Yoon et al., 2017). Table 4 lists studies that examined the impact of income levels

23
on the membership and usage of car-sharing. It should be stated that there is a different
perception of low, middle, or high income, and there are subgroups with distinct
behaviors/preferences. Therefore, for each study, the income range for the designated income
level (Low- or Moderate and Above-average or high), which has the largest share in the
distribution of members, is specified in the second column of Table 4. The unit currencies of
the countries listed in Table 4 have been converted to Euros per year for comparative purposes,
although incomes in different countries have different purchasing powers.
Table 4: The positive relationship of occupation and economic status groups on car-sharing
membership, usage, or attitudes.
Occupation
and economic
status groups
Average
household
income)
(euro/year)
%
distribution of
the members
Car-
sharing
service
type
Studied
impact
Specific
conditions
Geographic
area
References
Above-
average or
high-income
level
≥30000.0
77.0
Free-
floating
Membership
-
Netherlands
Winter et al.,
2017
≥30000.0
About 48.0
Free-
floating
Membership
-
Turin, Italy
Ceccato, 2020
≥30000.0
About 48.0
One-way
station-
based and
Free-
floating
Membership
-
Turin, Italy
Ceccato and
Diana, 2021
≥82836.0
Unspecified
One-way
station-
based and
free-
floating
Membership
-
Seattle. USA
Vinayak et
al., 2018
≥82836.0
Unspecified
One-way
station-
based and
free-
floating
Membership
-
Seattle. USA
Dias et al.,
2017
Net household
income ≥
24000.0
About 50.0
Round-
trip and
free-
floating
Membership
-
Berlin,
Germany
Kawgan-
Kagan, 2015
Canada: ≥
39767.0
USA: ≥
82836.0
50.0
Round-
trip, one-
way
station-
based
Membership
-
North America
Millard-Ball,
2005
≥ 82836.0
59.0
Station-
based
Membership
-
San Francisco
Bay Area,
USA
Clewlow,
2016
15000.0-
25000.0
Unspecified
One-way
station-
based
Willingness to
join
-
Greece
Efthymiou et
al., 2013
13255.0-
26512.0
19.0
Round-
trip
Membership
-
North America
Martin et al.,
2010
≥ 12240.0
About 16.0
Round-
trip
Membership
-
Beijing, China
Yoon et al.,
2017
41420.0-
62130.0
18.0
P2P
Membership
-
Portland, USA
Shaheen et
al., 2018
Low- or
moderate-
income level
17800.0-
44520.0
58.2
Station-
based
Willingness to
continue
membership
BEV service
Seoul, South
Korea
Kim et al.,
2015
15000.0-
25000.0
Unspecified
Station-
based
Willingness to
join
-
Athens,
Greece
Efthymiou
and Antoniou,
2014
Median
Household
income:
42420.0
Unspecified
Round-
trip
Membership
-
San Francisco,
USA
Cervero et al.,
2007
≤ 82840.0
68.0
Round-
trip
Membership
-
North America
Martin and
Shaheen,
2011a

24
Results from previous studies are somewhat mixed. In Salerno, Italy, nearly 80% of
employed users were inclined to use the e-car-sharing service (Cartenì et al., 2016). Car-sharing
members generally are from families where the number of employed people is above average,
and they are from high-income households in Turin, Italy (Ceccato and Diana, 2021).
Nonetheless, Martin and Shaheen (2011a) figured out that shared cars primarily served the
middle class in North America. Nevertheless, in the latter study, more than 20% of the members
of the shared cars earned $100,000 or more. In San Francisco, USA, the average annual income
of round-trip car-sharing members was $ 57,000, higher than the city average, primarily since
more than 90% worked in professional fields (Cervero and Tsai, 2004).
Similarly, some studies showed that members are mostly middle-to-higher-income in
North America (Brook, 2004; Martin et al., 2010; Millard-Ball, 2005). However, it should be
noted that Millard-Ball (2005) conducted an online survey of shared car members. The results
of this survey are likely to over-represent the individuals with a high-income level because they
are more inclined to use their personal computers. Shaheen et al. (2018) mentioned that P2P
shared car members generally earned slightly more than the US population. For the most part,
this result is general since P2P car-sharing, like many shared mobility systems, is built in large,
higher-income cities. Similarly, in a study by Winter et al. (2017), this sample shows more
educated people than the national average. The geographical limitations of this study could
explain this problem in a sample of selected cities located in the metropolis of the Randstad
region, which is more prosperous.
On the other hand, Kortum and Machemehl (2012) mentioned that families with higher
income levels are less inclined to choose shared cars. They probably prefer their vehicles.
Importantly, in this study income variable is insignificant. Hence, the direct relationship
between membership and income may not be between the mode share and income.
The probability of using e-car-sharing is higher among lower-income groups than high-
income individuals in Seoul. It may imply that the current economic advantages are
unsatisfactory for this group (Kim et al., 2015). Also, in San Francisco, car-sharing trips
declined as income levels raised (Cervero et al., 2007). It is significant to highlight that this
study used the income factor as a numerical variable. Nevertheless, income has been used as a
categorical variable in most other studies to make it more informative. This can help us identify
which income group most members belong to, compare income groups, and discover potential
non-linear relationships.
Similarly, Efthymiou and Antoniou (2014) suggested that low-to-middle-income
individuals are more willing to join the car-sharing program in Greece. In this study, median-
income respondents earning between € 15,000 and € 25,000 per year are more inclined to join
the car. This may show that lower-income individuals find station-based car-sharing more
expensive and prefer public transport or walking. Also, high-income individuals prefer to use
their vehicles. It should be noted that the presence of children seems to decrease car-sharing

25
use among families with low and middle earning levels (Dias et al., 2017). This could be
because of financial hardship and the complexity of children's activities and travel patterns.
Overall, the income of people who want to use a car subscription is above-average,
especially in a free-floating system. Indeed, it may not be easy to offer shared vehicles such as
free-floating car-sharing in low-income neighborhoods because it may not be profitable for
commercial operators. However, for people with lower-than-average incomes, car-sharing is
attractive. These people seem to think purchasing and maintaining a personal car is expensive.
However, they do require it for their causal travels. Therefore, it is likely that certain local
circumstances, such as the availability and attractiveness of other travel means like public
transport, may determine which social group tends to use shared cars.
Furthermore, it should be noted that the reasons high-income people are attracted to car-
sharing can differ from those of low-income people. In this regard, Millard-Ball (2005) noted
that individuals with various earnings stated various causes for utilizing shared cars. For
instance, people who earned between $ 10,000 and $ 20,000 a year (4% of the sample) looked
for trip comfort. People with incomes between $20,000 and $30,000 a year (7.7% of the
sample) demanded acceptable trip costs, needed to carry their belongings, and were reluctant
to use public transportation. People with income between $30,000 and $40,000 a year (11.3%
of the sample) looked for acceptable trip costs. Finally, people earning more than $ 75,000 a
year (35% of the sample) need a car for their destination and are looking for a low-cost means
of transport. This shows that middle- to upper-income members can also be cost-sensitive
people. Further, it is necessary to emphasize that their neighborhood’s shared car system may
not be conveniently provided.
2.1.4.1.5 Household size
Car-sharing users are in smaller households than the average (Ceccato and Diana, 2021;
Ceccato, 2020; Kortum and Machemehl, 2012; Millard-Ball, 2005). Table 5 lists studies that
showed a positive correlation between small household size and car-sharing use. In order to
clarify the meaning of small household size, for each study, the household size considered to
be small is specified in the first column of Table 5.
Table 5: The positive correlation between small household size and car-sharing membership,
usage, or attitudes.
Average
household size
Car-sharing service
type
Studied impact
Specific
conditions
Geographic
area
References
About 2.5
Free-floating
Usage
-
Austin, USA
Kortum and
Machemehl, 2012
About 2.4
Free-floating
Membership
Turin, Italy
Ceccato, 2020
Around 2.5
One-way station-based
and free-floating
Membership
-
Turin, Italy
Ceccato and Diana,
2021
About 2.0
Round-trip, one-way
Station-based
Membership
-
North America
Millard-Ball, 2005
1.8
Station-based
Membership
-
Portland, USA
Cooper et al., 2000
It is worth mentioning that if household income rises, the likelihood of buying a car-sharing
subscription increases (Clewlow, 2016; Dias et al., 2017); this is associated with the number
of employees in the house, a similar trend. However, the number of household members

26
negatively impacts shared car use (Ceccato and Diana, 2021). This can indicate that shared car
is utilized by employees living in low-size families. For example, in Portland, Oregon, the
household size of station-based car-sharing members was 1.8 people per household, while the
average city household was 2.23 people per household (Cooper et al., 2000). In Canada, the
probability of car-sharing members living with someone else was 71%. However, that figure
was 61% for US car-sharing members. Also, in North America, about 64% of members live
with at least another individual, with a household mean of 2.02. In addition, about a quarter of
families have children (Millard-Ball, 2005). Therefore, the car-sharing decline due to the
average household size increase is probably due to the more significant number of children in
larger families (Kortum and Machemehl, 2012). Because sometimes, the presence of children,
especially among low-and-middle-income households, can be accompanied by decreased
shared car membership. It is worth expressing that these results are based on only a few articles.
Therefore, more research is required to add strength to the results.
2.1.4.1.6 Marital status
Many single-person households use car-sharing systems in Austin, USA (Celsor and Millard-
Ball, 2007). Generally, the shared car is more appealing in places where the ratio of single-
parent households is high (Carroll et al., 2017; Coll et al., 2014). Table 6 lists studies on the
impact of being single on car use.
Table 6: The positive correlation between being single and car-sharing membership, usage,
or attitudes.
Car-sharing service
type
Studied impact
Specific conditions
Geographic area
References
Station-based
Intention to join car-sharing
-
Athens, Greece
Efthymiou and Antoniou,
2014
Round-trip
Membership
-
Dublin, Ireland
Carroll et al., 2017
Round-trip
Usage
-
USA
Celsor and Millard-Ball, 2007
Generally, married people are less inclined to utilize shared cars in Athens, Greece
(Efthymiou and Antoniou, 2014). This may be because a married couple may commute to
different workplaces, and both may use a personal car. Because using two shared cars or a
private car and car-sharing can be very costly for them. For example, the husband/wife can
take the wife/husband to the nearest public transport or workplace instead of the shared car.
It should be mentioned that only a few articles examine the impact of marital status on car-
sharing demand. Therefore, more studies are needed to understand its effects on car-sharing
demand, especially in free-floating and P2P services.
2.1.4.1.7 Presence of children
Some studies suggest that families with children are more inclined to opt for shared car schemes
(Carroll et al., 2017; Coll et al., 2014; Rotaris and Danielis, 2018; Sioui et al., 2013). Depending
on local conditions, this could be due to child seats in car-sharing vehicles. Indeed, some other
studies have suggested that the presence of children may be associated with reduced car-sharing
use (Kim et al., 2017; Kopp et al., 2015; Vinayak et al., 2018), especially among low- and
middle-income households (Dias et al., 2017). This may occur because of the more complex

27
travel-activity patterns created by children and also budget constraints. For instance, in Munich
and Berlin, Germany, most car-sharing members did not have children (Kopp et al., 2015).
Table 7 indicates a list of studies on the effect of the presence of children on car-sharing use.
Table 7: Effect of the presence of children on car-sharing membership, usage, or attitudes.
Presence of
Children
Car-Sharing Service
Type
Studied Impact
Specific
Conditions
Geographic area
References
Positive
Round-trip, one-way
station-based, free-floating,
P2P
Interested in car-
sharing
In rural areas
Friuli-Venezia
Giulia, Italy
Rotaris and
Danielis, 2018
Station-based and free-
floating
Membership
-
Montreal, Canada
Sioui et al., 2013
Station-based
Membership
-
Quebec City, Canada
Coll et al., 2014
Round-trip
Interested in car-
sharing
-
Dublin, Ireland
Carroll et al., 2017
Negative
Free-floating
Membership
-
Munich and Berlin,
Germany
Kopp et al., 2015
One-way station-based and
free-floating
Usage
-
Seattle. USA
Dias et al., 2017
One-way station-based and
free-floating
Usage
-
Seattle. USA
Vinayak et al.,
2018
Round-trip
Usage
-
Netherlands
Kim et al., 2017
Namazu et al. (2018) reported that the probability of being in the early stages of family
formation among the early users of one-way car-sharing is higher than among round-trip car-
sharing users. However, the survey data from this study is not enough to clarify whether users
of one-way shared cars become round-trip shared car users when they have children.
2.1.4.1.8 Vehicle ownership
In most cases, the mean number of cars in each family among the members of the car-sharing
systems is less than among non-members (Becker et al., 2017a; Catalano et al., 2008; Ceccato
and Diana, 2021; Ceccato, 2020; Cervero et al., 2007; Clewlow, 2016; De Luca and Di Pace,
2015; Efthymiou and Antoniou, 2014; Habib et al., 2012; Juschten et al., 2017; Namazu et al.,
2018; Nobis, 2006; Wang et al., 2012; Wang et al., 2017). Table 8 shows a list of studies
showing the positive correlation between the low level of vehicle ownership and the use of
shared cars.
Table 8: Positive correlation between low vehicle ownership and car-sharing membership,
usage, or attitudes.
Average household
vehicle ownership
(vehicle/household)
Car-sharing
service type
Studied
impact
Direction of
causation
Specific
conditions
Geographic
area
References
1.4
Free-floating
Membership
Exogenous
-
Turin, Italy
Ceccato, 2020
1.0
Free-floating
Usage
Exogenous
E-car-sharing
Germany
Burghard and
Dütschke, 2019
Average car per adult:
about 0.4
Free-floating
Membership
Exogenous,
Endogenous
-
Munich and
Berlin, Germany
Kopp et al.,
2015
1.1
Station-based
and free-
floating
Membership
Exogenous,
Endogenous
-
California, USA
Mishra et al.,
2019
0.1
Station-based
and free-
floating
Membership
Exogenous,
Endogenous
-
Montreal,
Canada
Sioui et al.,
2013
Households with one or
two vehicles
One-way
station-based
and free-
floating
Membership
Exogenous
-
Turin, Italy
Ceccato and
Diana, 2021

28
Average household
vehicle ownership
(vehicle/household)
Car-sharing
service type
Studied
impact
Direction of
causation
Specific
conditions
Geographic
area
References
Households with zero or
one vehicle
One-way
station-based
and free-
floating
Usage
Exogenous
-
Seattle. USA
Dias et al., 2017
0.4
Round-trip-
Membership
Endogenous
-
Vancouver,
Canada
Lempert et al.,
2019
1
Free-floating
About 0.8
Round-trip
Membership
Exogenous
-
Vancouver,
Canada
Namazu et al.,
2018
0.9
One-way
(Mainly free-
floating,
partially
Station-
based)
About 1.2
Round-trip,
free-floating,
and P2P
Membership
Exogenous
-
Switzerland
Juschten et al.,
2017
Unspecified
Station-based
Membership
Exogenous
-
Athens, Greece
Efthymiou and
Antoniou, 2014
About 0.6
Station-based
Membership
Exogenous
-
San Francisco
Bay Area, USA
Clewlow, 2016
About 0.5
Station-based
Membership
Exogenous
-
San Francisco,
USA
Ter Schure et
al., 2012
About 0.7
Station-based
Membership
Exogenous
-
Montreal,
Canada
Habib et al.,
2012
About 0.2
One-way
station-based
Membership
Exogenous
-
California, USA
Mishra et al.,
2015
Less than about 0.8
One-way
station-based
Membership
Exogenous
-
Salerno, Italy
De Luca and Di
Pace, 2015
About 0.7
Round-trip
Usage
Exogenous,
Endogenous
-
USA
Celsor and
Millard-Ball,
2007
About 0.2
Round-trip
Membership
Exogenous,
endogenous
-
North America
Martin et al.,
2010
Households with zero or
one vehicle
Round-trip
Membership
Exogenous
-
North America
Martin and
Shaheen, 2011a
Households with zero or
one vehicle
Round-trip
Membership
Exogenous,
Endogenous
-
San Francisco,
USA
Cervero et al.,
2007
0.3
Round-trip
Membership
Endogenous
-
San Francisco,
USA
Cervero and
Tsai, 2004
To clarify the meaning of low ownership level, for each study, the vehicle ownership range
considered a low level of vehicle ownership is specified in the first column of Table 8. It is
important to note that vehicle ownership, unlike the previously reviewed socio-economic
factors, can be seen as an exogenous variable (thus impacting car-sharing demand) and an
endogenous variable (since car-sharing might impact vehicle ownership levels). It is important
to distinguish the two opposite directions of causation from a transport policy viewpoint,
although the literature does not focus adequately on such aspects. Therefore, the fourth column
in Table 8 indicates whether vehicle ownership levels are considered exogenous, endogenous,
or (perhaps more realistically) a mix.
Some studies have shown that vehicle ownership affects car-sharing demand. For example,
in San Francisco in 2010, the average vehicle ownership for station-based car-sharing members
was 0.47 vehicles per household, and for non-members, 1.22 vehicles per household (Ter
Schure et al., 2012). The explanation that can be given is that most of the decline in vehicle
ownership seems to be related to shifting to walking, cycling, and transit and shortening the
average daily travel distance. Similarly, in Montreal, Canada, car-sharing members own fewer
private cars than average (Sioui et al., 2013). Besides, in the US, households without vehicles

29
or one vehicle have the highest rate of shared car use (Celsor and Millard-Ball, 2007).
Regardless of residential density, the high level of vehicle ownership adversely influences one-
way station-based and free-floating shared car usage (Dias et al., 2017). Probably, it is more
comfortable and cost-effective for individuals to use personal cars than shared cars.
In general, the mobility behavior of car-sharing system members is more sustainable, and
they are more multimodal than non-members (Becker et al., 2017a; Clewlow, 2016; Costain et
al., 2012; Wang et al., 2017). Car-sharing is generally accepted by people who reside in families
with fewer personal vehicles than non-members (Chicco et al., 2020). In this regard, Clewlow
(2016) figure out that in city regions, members of station-based car-sharing own fewer cars
(0.58) than non-members (0.96). It was shown that car-sharing system members have only
made up 41.5% of their private cars’ travels, but this figure is 61.8% for non-members. Also,
car-sharing members have carried out about 15% of their travels in transit and around 35% of
their travels on foot. However, these figures for non-members are 10.3% and 23.0%,
respectively. Hence, car-sharing is linked to multimodal travel behavior. This effect looks
greater for the station-based shared systems members (Namazu et al., 2018). Also, shared car
members are more inclined to own cars with low carbon footprints (Kawgan-Kagan, 2015).
Also, they are more inclined to have more sustainable car technologies. The portion of Ev's use
is remarkably more among car-sharing members. Besides, about one-fifth of cars owned by
car-sharing members were hybrid, plugin hybrid, or BEVs, while the diffusion rate of such
vehicles among non-members is halved (Clewlow, 2016). This may indicate a possible link
between membership in car-sharing and environmental attitudes.
In a study by Chicco et al. (2020), it was noted that in Frankfurt, Germany, people who
chose both free-floating and station-based programs had less private car ownership than people
who utilized only the free-floating service. Further, it was stated that in the Brussels Capital
Region, the round-trip service members have five times fewer private cars than free-floating
service members. Around 62% of round-trip car-sharing system members in the USA are from
households that did not have a private car when joining car-sharing, and 31% of members had
only one car. Therefore, more than 90% of them did not have more than one car (Martin and
Shaheen, 2011a).
Some studies have indicated the effects of shared cars on car ownership. For example, in
Montreal, Canada, car usage by people, who did not have a vehicle and used shared cars more
than 1.5 times a week, was 25% lower than vehicle owners. This difference arises with a
reduction in the frequency of car-sharing services usage (Sioui et al., 2013). This confirms the
remarkable effect of car-sharing usage. Furthermore, round-trip car-sharing service usage
sometimes decreases car ownership and use (Celsor and Millard-Ball, 2007). In North America,
around one-third decline in the mean car kilometers traveled before and after joining the round-
trip car-sharing program was observed. This figure was 6468 km per year for the former and
4729 km per year for the latter (Martin and Shaheen, 2011b). This reduction of about 1740 km
per year means a 27% reduction in the driving distance before and after. In North America,
round-trip car-sharing members’ vehicle ownership dropped dramatically from around 0.47
cars per household to about 0.24 cars per household (Martin et al., 2010). Hence, the car-

30
sharing service can facilitate a reduction in ownership of household vehicles as this service
dramatically eliminates the need for a personal vehicle to complete travel. That way, car-
sharing can only provide a car to a member if needed. Out of every 25 households joining
round-trip car-sharing, six would shed off their private car within two years in San Francisco
(Cervero and Tsai, 2004). The comfort of having access to a fleet of cars on demand may
encourage some car owners to dispose of their second vehicles and give up car ownership
altogether.
Similarly, Becker et al. (2017a) indicated that half of the comparison group members used
their vehicles at least once weekly. However, it is 14% for free-floating shared car system
members and 4% for station-based shared system members. It seems that members of different
shared car system types belong to different households. Moreover, the motivation of the round-
trip members is more for financial and environmental reasons. On the other hand, one-way
shared car members are more motivated with more convenience and safety. In addition,
members of one-way car-sharing consider car-sharing as an alternative to ride-hailing systems
like Uber or Lyft. Round-trip members, however, see the shared car as a substitute for car
ownership and a way to travel out of the city (Lempert et al., 2019).
Looking at different geographic areas, if station-based car-sharing programs were available
in China, a small percentage (11%) of households with a private car would tend to shed one.
This ratio is lower than that of previous European and North American research. However,
those who want to buy a private car in the short term, within one year to three years, consider
car-sharing because most of them tend to give up their purchase plans (Wang et al., 2012).
Therefore, car-sharing in China seems to be more effective in preventing the purchase of
vehicles than car-shedding. Car-sharing, especially free-floating services, may significantly
influence postponing the purchase of additional private cars in Italy. However, in the Brussels
Capital Region, members of free-floating car-sharing services did not necessarily see the
service as a replacement for their private car but as a supplement (Chicco et al., 2020). In this
regard, it should be stated that free-floating shared car members are more likely to agree that
the personal vehicle is a symbol of status (Burghard and Dütschke, 2019).
The free-floating shared car program influenced the car ownership of 37% of users in
London. Of this 37%, most users (83%) reported not wanting to purchase a private vehicle after
car-sharing. Furthermore, 11% stated that they had not used their vehicle in the previous three
months, and 6% indicated that they would sell their vehicle within the next three months (Le
Vine and Polak, 2019). However, 63% of members stated that the car-sharing system did not
influence their car ownership status. Some concerns can be raised because Le Vine and Polak
(2019) surveyed users only three months after introducing the free-floating system in London.
Users may change their minds after a while. Hence, these results may not reflect their actual
long-term behavior. Also, most of that 37% of users probably did not own a private car.
There seems to be a complex two-way relationship between shared car membership and
owning a car. For instance, in a survey by Martin et al. (2010), approximately 30% of
respondents noted that they had joined car-sharing to throw away their cars or avoid purchasing
an extra car. This highlights the influence of shared cars on vehicle ownership status. This

31
group can be extended to suburban residents who do not access shared cars in their
neighborhoods but utilize car-sharing when visiting city centers or workplaces. On the other
hand, about 50% of respondents stated that they did not have a vehicle and had joined a shared
car program to access the vehicles. This determines that the strength of the relationship is in
the opposite direction. There may be a hypothesis that car-sharing affects increased driving and
travel but does not reduce vehicle ownership. The second group of members joins the shared
cars to reduce car ownership; however, further research is required to address such
heterogeneity.
Some studies, such as Martin et al. (2010) and Firnkorn and Müller (2012) on the impact
of car-sharing causality, have been conducted according to surveys of shared car members. The
research addressed the two-way relationship between car ownership and car-sharing.
Therefore, they try to control the reverse causality bias by examining the number of people’s
cars before registering in the shared car program and then the number of their cars after
registration. The research did not evaluate impacts by comparing the changes with a
comparison group. Instead, they assessed the impacts by asking respondents to describe their
decision to car-shed and sign-up for car-sharing. For instance, in a study by Firnkorn and
Müller (2012), car-sharing members were asked to explain whether their decision to eliminate
or ignore future car purchases was taken because of using shared car programs or other reasons.
Some studies have inferred causal impacts by comparing the trip behavior of members with
non-members (Kopp et al., 2015; Sioui et al., 2013).
Moreover, to draw causal inferences, Cervero et al. (2007) compared the trip behavior of
the members of shared car programs with those of individuals who requested to be part of a
car-sharing scheme but were not yet (control group). It turned out that members of round-trip
car-sharing avoid using personal cars almost 12% more than non-members. A decrease in car
possession can accompany membership and a decline in car possession with more shared car
travels.
Mishra et al. (2015) applied a survey to investigate the effects of shared cars on trip
behavior. Propensity score matching was utilized to control the self-selection bias resulting
from the observed differences. Each member has matched non-members with the same person
and family demographics and lives vicinities with an analogous built environment. Vehicle
ownership of members is significantly less than that of non-members. This difference also
increases with the desire to register a car-sharing. However, there is a simultaneity bias in this
study. Also, there is possibly the self-selection bias that differences in unobserved features may
cause. Hence, this study cannot claim that car-sharing can cause the observed differences in
trip behavior between matched pairs.
Mishra et al. (2019) estimated the car-sharing impact on car ownership and current
members’ trip behavior using the California household travel survey database. However, in
this study, the surveys have not explored the features of trip behavior, particularly the
chronology of events that might result in inverse causation.

32
To sum up, round-trip shared car service members may follow a more efficient and
sustainable lifestyle than the one-way shared car system members. Sometimes, this difference
can be significant, especially in China, where the effect of choosing car-sharing is more to
prevent purchasing a new car than to reduce car ownership. For instance, a study conducted in
Beijing, China, indicated that car ownership positively affects the number of one-way trips and
negatively influences the round-trip travel numbers (Yoon et al., 2017). Generally, people
attracted to the station-based shared car program have less vehicle ownership than those
attracted to the free-floating shared car program. Besides, station-based shared car members
can decrease vehicle ownership more than free-floating shared car members. Also, it should
be stressed that the mean number of cars per family for car-sharing members in North America
seems lower than in Europe.
Generally, most studies have focused on the effect of vehicle ownership on shared cars.
However, further research on this two-way relationship is needed to have a deep insight into
the direction of causation between shared cars and car ownership and consequently assess the
sustainability of shared cars.
2.1.4.2 Trip-related characteristics
Trip-related characteristics such as travel time, departure time, travel purpose, and travel
distance can play an essential role in the car-sharing demand rate.
2.1.4.2.1 Travel time
Whenever car-sharing users save significantly on trip time, they are willing to pay market
prices for these advantages (Cervero, 2003; Carroll et al., 2017). The longer the travel time, the
less satisfaction (Catalano et al., 2008; Efthymiou et al., 2013). Time pressure has an adverse
effect on encouraging people to choose a shared car in the Netherlands (Kim et al., 2017c).
Private cars are generally less time-consuming than car-sharing systems. Nonetheless, car-
sharing services can outperform the subway, buses, and walking in terms of travel time
(Martínez et al., 2017). Table 9 documents the positive impact of shorter travel time on car-
sharing usage.
Table 9: The positive correlation between shorter travel time and car-sharing usage.
Car-sharing service type
Geographic area
References
One-way station-based
Greece
Efthymiou et al., 2013
One-way station-based
Palermo, Italy
Catalano et al., 2008
Round-trip
Netherlands
Kim et al., 2017c
Round-trip
Dublin, Ireland
Carroll et al., 2017
Round-trip
San Francisco, USA
Cervero, 2003
2.1.4.2.2 Travel distance
According to Li (2019), the car-sharing choice can vary depending on the travel distance. For
example, the value of travel time savings (VTTS) for car-sharing in China is about $3.3 per
hour for middle-distance travel (2 km to 5 km) and $12.2 per hour for long-distance travel
(beyond 5 km). Hence, VTTS typically increases with travel length. Besides, to enhance car-
sharing service usage, policies should focus on saving users' travel time for longer trips and
saving users' travel costs for short trips. For example, the propensity for choosing car-sharing

33
rises with trip length in Lisbon, Portugal (Martínez et al., 2017). It means that the longer the
trip, the more likely people are to choose a car-sharing system. Similarly, individuals interested
in car-sharing services have long commutes in Shanghai, China (Wang et al., 2012). However,
in Toronto, car-sharing has played a role in increasing short-distance auto urban trips (Costain
et al., 2012). Besides, car-sharing members usually have shorter commutes than most
individuals living in the same area. Households living near their workplace mostly use the car-
sharing program (Martin and Shaheen, 2011a). Table 10 shows the positive effect of different
trip distance ranges on car-sharing usage.
Table 10: The positive relationship between different trip distance ranges and car-sharing
usage.
Trip distance ranges
Car-sharing service type
Geographic area
References
Long
One-way station-based
Lisbon, Portugal
Martínez et al., 2017
One-way station-based
Taiyuan, China
Li, 2019
Round-trip
Shanghai, China
Wang et al., 2012
One-way station-based
Taiyuan, China
Li, 2019
Short
Round-trip
North America
Martin and Shaheen, 2011a
Round-trip
Toronto, Canada
Costain et al., 2012
2.1.4.2.3 Departure time
Car-sharing systems are commonly utilized to travel during off-peak hours or weekends when
transport services are inadequate and have low traffic (Costain et al., 2012). Their use is also
related to trip purposes because shopping and leisure or social trips are often made during off-
peak hours. Also, car-sharing systems are generally not utilized during peak periods (Cervero,
2003). However, there was an insignificant correlation between peak-hour travel and demand
for car-sharing services in Lisbon, Portugal (Martínez et al., 2017). It is important to stress that
potential members do not utilize car-sharing services for systematic workday travel, even if the
system is appropriate for urban trips on congested roads (short-distance and high-duration trips)
(Ceccato, 2020). Table 11 covers the positive effect of travel on the rate of car-sharing use on
weekends, during off-peak hours, or in the morning.
Table 11: The positive correlation between weekend traveling, off-peak hours, or in the
morning and car-sharing usage.
Car-sharing service type
Geographic area
References
Free-floating
Turin, Italy
Ceccato, 2020
Round-trip
Toronto, Canada
Costain et al., 2012
Round-trip
San Francisco, USA
Cervero, 2003
2.1.4.2.4 Trip purpose
Car-sharing systems are more utilized for social, recreational, and personal business trips than
non-discretionary trips such as trips to school or work in San Francisco (Cervero, 2003). The
most common purpose of car-sharing users' travel is business activities in China (Wang et al.,
2017). More than 84% of users traveling for non-working purposes were satisfied with car-
sharing services in Salerno, Italy (Cartenì et al., 2016). Also, car-sharing is commonly utilized
for non-compulsory trips such as shopping and leisure trips (Martin and Shaheen, 2011a; Kim
et al., 2015). Users who do not have a car utilize the One-way car-sharing system to allow

34
people to shop less, go to grocery stores less, and spend less time shopping (Le Vine, Adamou,
and Polak, 2014).
Moreover, users who did not own a car were more likely to opt for Free-floating car-sharing
systems for shopping purposes because the Free-floating cargo capacity system helps users
carry bulky items (Le Vine and Polak, 2019). Finally, users are more likely to utilize BEV car-
sharing for leisure trips than for commuting travels (Jin et al., 2020). Table 12 sets out two
main trip purpose groups' impact on car-sharing use.
Table 12: Impact of different trip purpose groups to use car-sharing.
Trip purpose groups
Impact
Car-sharing
service type
Geographic
area
References
Social, Recreational, and Personal Business Trips,
Shopping Trips, Non-working Trips, Non-
commuting Trips
Positive
Effect
Free-floating
China
Wang et al., 2017
Free-floating
London, England
Le Vine and Polak,
2019
One-way
station-based
London, England
Le Vine, Adamou, and
Polak, 2014
One-way
station-based
Salerno, Italy
Cartenì et al., 2016
One-way
station-based
Beijing, China
Jin et al., 2020
Station-based
Seoul, South
Korea
Kim et al., 2015
Round-trip
North American
Martin and Shaheen,
2011a
Round-trip
San Francisco,
USA
Cervero, 2003
Non-discretionary trips such as travels to school or
work
Negative
Effect
Round-trip
San Francisco,
USA
Cervero, 2003
2.1.4.3 Car-sharing characteristics
One of the most significant factors affecting car-sharing demand is the travel mode attributes,
such as travel cost and comfort (Carroll et al., 2017). In a study considering the switching from
private cars to EV car-sharing systems, the trip cost was the primary determinant of the
selection process. In comparison, trip time changes did not significantly change the probability
of switching from private cars to EV car-sharing systems (Cartenì et al., 2016).
The effect of the main car-sharing characteristics is reviewed in the following.
2.1.4.3.1 Travel cost
Trip cost is a significant factor in users' car-sharing choice behavior (Catalano et al., 2008;
Lamberton and Rose, 2012; Carroll et al., 2017). Lower fares and more electric car supply can
increase students who use car-sharing from 2% to 10-15% in Italy (Rotaris et al., 2019).
Although travel time is statistically significant, travel cost had a much larger effect on car-
sharing choices than travel time in Salerno, Italy (Cartenì et al., 2016). Similarly, travel costs
were determined as important factors along with access time to car-sharing parking spaces,
travel frequency, car availability, travel type (home-based), gender, and age in Salerno.
Besides, changes in the car-sharing trip's cost had a much more significant impact on the
likelihood of choosing a carpool than the probability of selecting a bus and private car (De
Luca and Di Pace, 2015).

35
In both Round-trip and One-way travel, the cost gap (the cost of the original transport mode
minus the car-sharing cost) significantly influences car-sharing choice. The consistency of the
cost gap impact in One-way and Round-trip models emphasizes the significance of competitive
fares for successful car-sharing systems (Yoon et al., 2017). More than 25% of those interested
in the car-sharing program stated that if this system is reasonably priced, car-sharing usage will
be considered by people in Beijing (Shaheen and Martin, 2010). Table 13 lists those studies
documenting the positive impact of the low travel cost on car-sharing use.
Table 13: The positive correlation between the low travel cost and car-sharing use.
Car-sharing service type
Geographic area
References
One-way station-based
Palermo, Italy
Catalano et al., 2008
One-way station-based
Beijing, China
Shaheen and Martin, 2010
Round-trip
USA
Lamberton and Rose, 2012
One-way station-based
Salerno, Italy
De Luca and Di Pace, 2015
One-way station-based
Salerno, Italy
Cartenì et al., 2016
One-way and Round-trip
Beijing, China
Yoon et al., 2017
Round-trip
Dublin, Ireland
Carroll et al., 2017
Round-trip, One-way station-based, Free-floating
Rome and Milan, Italy
Rotaris et al., 2019
2.1.4.3.2 Travel comfort
Travel comfort can affect people's car-sharing choices, but only one study considered comfort
an important factor in the USA's free-floating system (Schaefers, 2013).
2.1.4.4 Built environment and land use
Built environment and land use characteristics such as accessibility to car-sharing systems,
fleet size, fleet age, and land use are the last factors considered in this review.
2.1.4.4.1 Land use
In general, many car-sharing members frequently use public transportation and live in medium
to high-density areas (Cervero, 2003; Shaheen and Rodier, 2005; Burkhardt and Millard-Ball,
2006; Kortum and Machemehl, 2012; Kopp et al., 2015; Wagner et al.,2016; Dias et al., 2017;
Namazu et al., 2018). Higher regional population levels are associated with more extended
membership periods in the car-sharing program (Habib et al., 2012). According to Hu et al.
(2018), an area with greater road density, a higher population density, or mixed land use is
associated with higher car-sharing system use rates. Besides, stations around shopping malls,
colleges, and transit hubs can attract more users to car-sharing. However, car-sharing stations
are often oversupplied in transportation hubs. Also, car-sharing is more effective in areas with
limited access to subway services. Millard-Ball (2005) stated that car-sharing is mainly
concentrated in urban cores, and about 95% of the users are observed in these settings. A
suitable environment for pedestrians, high density, and a combination of parking pressures and
uses can contribute to car-sharing service success. It is important to consider that a private
parking lot near the house severely negatively affects car-sharing system usage rates (Juschten
et al., 2017; Ceccato and Diana, 2021). Table 14 presents the influence of two different land-
use patterns on car-sharing usage.
Table 14: Impact of different land-use patterns to use car-sharing.
Land use patterns
Impact
Car-sharing service
type
Geographic area
References
Living in Urban Cores, Medium to High Densely
Populated Areas, Mix Land Use, Areas Where
Positive
Free-floating
Berlin, Germany
Wagner et
al.,2016

36
Land use patterns
Impact
Car-sharing service
type
Geographic area
References
Public Transportation Does Not Provide Service,
Stations Around Shopping Malls, Colleges, And
Transit Hubs
Free-floating
Munich and
Berlin, German
Kopp et al., 2015
Free-floating
Based,
Switzerland
Becker et al.,
2017a
Free-floating
Austin, USA
Kortum and
Machemehl, 2012
One-way station-
based and Free-
floating
Seattle. USA
Dias et al., 2017
Round-trip, One-way
station-based, Free-
floating
North America
Millard-Ball,
2005
Round-trip, One-way
(Mainly Free-
floating, partially
Station-based)
Vancouver,
Canada
Namazu et al.,
2018
Station-based
Montreal, Canada
Habib et al., 2012
Station-Based
Shanghai, China
Hu et al., 2018
Round-trip
USA and Canada
Burkhardt and
Millard-Ball,
2006
Round-trip, One-way
station-based,
Business-to-Business
(B2B)
San Francisco
Bay Area, USA
Shaheen and
Rodier, 2005
Round-trip
San Francisco,
USA
Cervero, 2003
Private Parking Lot Near the House
Negative
One-way station-
based and Free-
floating
Turin, Italy
Ceccato and
Diana, 2021
Round-trip, Free-
floating, and Peer-to-
Peer
Switzerland
Juschten et al.,
2017
2.1.4.4.2 Accessibility
Ease of access is considered an important factor in car-sharing (Ciari and Axhausen, 2012).
Also, there is an interrelationship between Station-based car-sharing systems and public
transportation accessibility (Stillwater et al., 2008). In general, access to stations in terms of
the distance between home/work and the nearest station is a dominant factor in joining a car-
sharing program (Brook, 2004; Zheng et al., 2009; Costain et al., 2012). In addition, availability
significantly influences the likelihood of using car-sharing (Kim et al., 2017b).
Most car-sharing system members have access to services from less than 1 km in Toronto,
Canada (Costain et al., 2012). Increasing the number of stations within a 5 km radius of the
household raises the likelihood of car-sharing membership in Switzerland (Juschten et al.,
2017). Wider streets and regional rail access lead to lower demand rates in average monthly
car-sharing usage hours. In contrast, the exclusive availability of light rail can lead to higher
demand (Stillwater et al., 2008). Enacting active policies to limit private transport usage could
raise car-sharing use by up to 10% in Palermo, Italy (Catalano et al., 2008). Table 15 details
the impact of different accessibility conditions on car-sharing usage.
Table 15: Impact of different accessibility conditions to use car-sharing.
Accessibility condition
Impact
Car-sharing service
type
Geographic
area
References
Less Distance Between Home/Work and The Nearest
Station, Shared Car Availability, High Number of Car-
Positive
Round-trip, Free-
floating, and Peer-to-
Peer
Switzerland
Juschten et al.,
2017

37
Accessibility condition
Impact
Car-sharing service
type
Geographic
area
References
Sharing Stations, Higher Rates of Only Light Rail
Availability, Limiting Private Transport Usage
One-way station-based
Palermo, Italy
Catalano et al.,
2008
Round trip and One-
way station-based,
Business-to-Business
(B2B)
North America
Brook, 2004
Round-trip
Toronto,
Canada
Costain et al.,
2012
Round-trip
Madison, USA
Zheng et al.,
2009
Round-trip
USA
Stillwater et al.,
2008
Buyer-to-Consumer
Netherlands
Kim et al.,
2017b
More street width and regional rail access
Negative
Round-trip
USA
Stillwater et al.,
2008
2.1.4.4.3 Size and age of stations
Car-sharing station size substantially affects the availability and usage of car-sharing stations
in Montreal, Canada. Also, larger stations have larger catchment basins than smaller ones and
provide more vehicle options (De Lorimier and El-Geneidy, 2013). Although increasing the
number of cars at stations does not necessarily affect member subscriptions, monthly usage
increases (Habib et al., 2012). In addition, older car-sharing stations lead to higher demand for
car-sharing systems (Stillwater et al., 2008). Table 16 lists papers assessing the positive effect
of larger and older stations on car-sharing usage.
Table 16: The positive correlation between larger and older stations and car-sharing usage.
Car-sharing service type
Geographic area
References
Station-based
Montreal, Canada
De Lorimier and El-Geneidy, 2013
Station-based
Montreal, Canada
Habib et al., 2012
Round-trip
USA
Stillwater et al., 2008
2.1.4.5 Attitudinal effects (subjective factors)
Many studies have been done on the decision to use car-sharing for daily mobility. However,
there are more opportunities to increase the behavioral realism of shared mobility choice
models, and examining the impact of personal attitudes on mode choice decisions is one
potential path. Regarding car-sharing choices, only a few recent studies have investigated the
potential impact of a limited range of attitudinal factors, reviewed in the following subheads.
2.1.4.5.1 User satisfaction
User satisfaction with car-sharing services is considered an influential factor in their usage
rates. People's satisfaction with their current travel patterns can significantly impact their
intention to join a car-sharing program (Efthymiou and Antoniou, 2016; Kim et al., 2017b).
Also, people who used car-sharing were satisfied with the service and wanted to use it in the
future in Turin, Italy (Ceccato, 2020). Especially on congested streets, car-sharing appeared to
be attractive for city travel. Besides, the expectation of perceived effort (e.g., degree of ease
associated with use) could be one of the most influential psychological elements which can
indicate the intention to use Business-to-Business (B2B) services (Fleury et al., 2017). Table
17 lists those studies that documented the positive effect of user satisfaction on car-sharing
usage.

38
Table 17: The positive correlation between user satisfaction and car-sharing usage.
Car-Sharing Service Type
Geographic area
References
Free-floating
Turin, Italy
Ceccato, 2020
One-way station-based
Athens, Greece
Efthymiou and Antoniou, 2016
Buyer-to-Consumer
Netherlands
Kim et al., 2017b
Business-to-Business (B2B)
France
Fleury et al., 2017
2.1.4.5.2 Service awareness, environmental concerns, and social impact
Generally, individuals familiar with the car-sharing scheme are more likely to use it (Duan et
al., 2020). Also, people aware of car-sharing are more likely to forgo private car purchases
(Wang et al., 2017). The car-sharing choice was correlated with the attitude toward "Advocacy
of car-sharing service" in Taiyuan, China (Li, 2019). Car-sharing use is also positively related
to pro-environmental and privacy attitudes (Kim et al., 2017c). Pro-environmental and pro-
technology attitudes positively correlate with car-sharing systems' perceived advantages
(Acheampong and Siiba, 2020). The frequency of car-sharing usage rates was influenced by
attitudes such as pro-environmental and neo-urban lifestyle preferences and socio-interactions
(for example, people's behavior depends on their loved ones' behavior) in Seattle, USA
(Vinayak et al., 2018). Furthermore, the social impact of car-sharing choices is important. The
degree of social impact varies according to social relationships' strength in individuals (Kim et
al., 2017a).
Table 18 illustrates the positive effect of high levels of environmental concerns and social
impact on car-sharing use.
Table 18: The positive correlation between the high level of environmental concerns and the
importance of social impacts and car use.
Car-sharing service type
Geographic area
References
Free-floating
China
Wang et al., 2017
One-way station-based and Free-floating
Seattle. USA
Vinayak et al., 2018
Station-based and Free-floating
Ghana, Sub-Saharan Africa.
Acheampong and Siiba, 2020
One-way station-based
Taiyuan, China
Li, 2019
One-way station-based
Shanghai, China
Duan et al., 2020
Round-trip
Netherlands
Kim et al., 2017a
Round-trip
Netherlands
Kim et al., 2017c
2.1.4.5.3 User's habits
People's habits can significantly affect their intention to use car-sharing (Efthymiou and
Antoniou, 2016; Kim et al., 2017b; Zhou et al., 2020). Commuters need the right motivation to
break the habits that may exist for a considerable period (Carroll et al., 2017). Hence, it is
important to consider the users' habits to estimate the car-sharing demand. Members' activity
in the last four months has affected the users' behavior in the current month in Montreal, Canada
(Morency et al., 2012). Table 19 reflects the positive impact of experience on car-sharing usage.

39
Table 19: The positive correlation between previous experience and car-sharing usage.
Car-sharing service type
Geographic area
References
Station-based
North America
Morency et al., 2012
One-way station-based
Athens, Greece
Efthymiou and Antoniou, 2016
Round-trip
Dublin, Ireland
Carroll et al., 2017
Buyer-to-Consumer
Netherlands
Kim et al., 2017b
Peer-to-Peer, Buyer-to-Consumer
Australia
Zhou et al., 2020
2.1.4.5.4 Private car status symbol
The car-sharing choice is correlated with perceptions of the car's symbolic value (Kim et al.,
2017c). Around 13% of the Free-floating car-sharing system's users concur with the statement
that the private car is a status symbol, while only 6% of users of the Station-based car-sharing
programs agree with it in Based, Switzerland (Becker et al., 2017a). An exploratory study on
citizens' acceptance of car-sharing in Beijing, China, was conducted by Shaheen and Martin
(2010). The results indicated that only 11% of the total sample cited the private car as a status
symbol, which probably indicates that mobility is a priority for most Beijing people rather than
property (Shaheen and Martin, 2010). Table 20 lists studies documenting the negative impact
of private cars on car-sharing usage as a status symbol.
Table 20: The negative correlation between private car symbol status and car-sharing usage.
Car-sharing service type
Geographic area
References
Station-based and Free-floating
Based, Switzerland
Becker et al., 2017a
One-way station-based
Beijing, China
Shaheen and Martin, 2010
Round-trip
Netherlands
Kim et al., 2017c
2.1.4.5.5 Sense of ownership
According to Paundra et al. (2017), psychological ownership refers to people's possessive
feelings about objects, whether the object legally belongs to them or not. The sense of
ownership can affect car-sharing usage. Also, low psychological ownership may lead to a
higher preference for a shared car under certain conditions. Besides, the price effect is less
pronounced for individuals with high psychological ownership. Due to their strong sense of
ownership over the target objects, such as cars, they prefer private cars to shared cars,
regardless of their low price. Table 21 shows the effect of the sense of ownership on car-sharing
usage.
Table 21: The positive correlation between sense of ownership and car-sharing usage.
Sense of Ownership Level
Impact
Car-Sharing Service Type
Geographic area
References
Low psychological ownership
Positive effect
Free-floating
Netherlands
Paundra et al., 2017
High psychological ownership
Negative effect
Free-floating
Netherlands
Paundra et al., 2017
2.1.5 Interaction effects among different factors
The previous section analyzed the effect of each factor on car-sharing demand. However,
interaction effects are expected and have been studied in the literature. Therefore, this section
focuses on the main ones documented in the literature. Table 22 shows a matrix mentioning

40
those papers that explicitly studied the interactions between two specific factors concerning
car-sharing use. Such interactions are then described in the following.
A study by Kawgan-Kagan (2015) indicated that females usually traveled shorter distances
than males. Early female adopters of car-sharing systems were likelier to use BEVs than
vehicles with internal combustion engines. They evaluated BEVs' performance positively,
especially in Free-floating car-sharing systems. When utilizing the charging station, 40 % of
females experienced a positive attitude, while even 20% of males did not state so. Dias et al.
(2017) suggested that children's presence in households without high income adversely
affected car-sharing usage rates in Seattle, USA. Rotaris and Danielis (2018) noted that in rural
areas, unlike metropolitan areas where car-sharing was common among professionals, car-
sharing programs were used mainly by the unemployed or students. Also, it was mentioned
that car-sharing system usage in rural areas was more common for non-commuting and longer
trips in rural areas. Moreover, Lamberton and Rose (2012) argued that the price was a
significant factor in selecting a shared car system. The primary concern was to profit from
selecting the shared vehicle for individuals with low psychological ownership. According to
Li (2019), the user's willingness to use the car subscription increased with increasing travel
distance in cold weather. In addition, when car-sharing was faced with a trade-off between time
and cost, travelers were more concerned with saving travel costs on shorter trips and saving
travel time on longer trips. Moreover, a car-sharing service for shorter trips was preferred for
non-commuting trips. While in the case of longer trips, it was highly preferred for commuting
trips. In addition, Wang et al. (2017) mentioned that individuals who knew better about the car-
sharing program, male users, and people with higher income levels accepted high prices.
Besides, Kim et al. (2015) noted that the members of electric car-sharing systems were likely
to retain their membership program mainly for non-compulsory trips. However, there was little
chance to change their car ownership behavior.

41
Table 22: Interactions matrix between factors on the use of car-sharing.
NB:“+”: positive interaction
“-”: negative interaction.
Lower travel
distance
Battery
electric
vehicles
(BEVs)
Low- and
middle-
income
households
Rural area
Lower travel
cost
Cold weather
Lower travel
time
Non-
commuting
trips
Commuting
trips
Male
Higher-
income levels
Familiarity
with the car-
sharing
program
Female
+
(Kawgan
-Kagan,
2015)
+
(Kawgan
-Kagan,
2015)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Presence of
Children
-
-
–
(Dias et
al.,
2017)
NA
NA
NA
NA
NA
NA
NA
NA
NA
Unemployed
or Students
NA
NA
NA
+
(Rotaris
and
Danielis
, 2018)
NA
NA
NA
NA
NA
NA
NA
NA
Low
psychologica
l ownership
NA
NA
NA
NA
+
(Lambe
rton
and
Rose,
2012)
NA
NA
NA
NA
NA
NA
NA
Short travel
distance
NA
NA
NA
NA
+
(Li,
2019)
NA
NA
+
(Li,
2019
)
NA
NA
NA
NA
Long travel
distance
(beyond 5
km)
NA
NA
NA
+
(Rotaris
and
Danielis
, 2018)
NA
+
(Li, 2019)
+
(Li,
2019
)
NA
+
(Li,
2019
)
NA
NA
NA
High travel
cost
NA
NA
NA
NA
NA
NA
NA
NA
NA
+
(Wan
g et
al.,
2017)
+
(Wan
g et
al.,
2017)
+
(Wang et al.,
2017)
Non-
commuting
trips
NA
+
(Kim et
al., 2015)
NA
+
(Rotaris
and
Danielis
, 2018)
NA
NA
NA
NA
NA
NA
NA
NA

42
2.1.6 Summary
2
This study's key conclusions are reported in the following lists, separately considering the five
factors.
The effect of different sociodemographic factors is summarized in the following list.
Besides, to assess the corresponding level of support in the literature, the number of articles
used to claim each result for the socio-demographic factors part is listed below.
• Gender: car-sharing seems to be accepted by both males and females (4 articles), even
if there is much attraction for potential female members (3 articles); males are more
likely to travel shorter distances and more frequently (1 article).
• Age: most car-sharing members/users are young (23 articles), typically in their mid-20s
to mid-30s (12 articles).
• Education level: being attracted to car-sharing may be based on a certain level of social
awareness, not strictly an economic decision (5 articles); most people looking to choose
car-sharing seem to have a four-year college degree or higher (14 articles), especially a
postgraduate or advanced degree (4 articles); beyond car-sharing membership, a high
level of education can also increase the utilization of car-sharing (20 articles).
• Occupation and economic status: most shared car members earn more than non-
members, and most are employed (12 articles); car-sharing members with middle to
upper-income levels can also be cost-sensitive people (1 article).
• Marital status: car-sharing is attractive in places where the proportion of single-parent
households is high (5 articles).
• Car ownership: the mean number of cars per family for car-sharing system members is
lower than for non-members (21 articles); there is a complex two-way relationship
between car ownership status and shared car demand (6 articles); car-sharing in China
seems to be more effective in preventing the purchase of vehicles than car-shedding (1
article).
However, several interaction effects between different socio-demographic factors have
been detected. The most important ones are the following:
• Between age and economic status: an older age (55 years or older) of people living in
households without high income negatively affects the propensity to join a car-sharing
scheme (1 article).
• Between age, marital status, and car-ownership status: car-sharing with Evs has a
special added attraction for young couples with no private car (1 article). The same is
true for young people who start a family and use car-sharing to complement their
private car trips (1 article).
• Between occupation status and household size: shared cars are more utilized by
employees living in low-size families (1 article).
2
Most of the contents of the present section/appendix have been published in Amirnazmiafshar, E., & Diana, M.
(2022). A review of the socio-demographic characteristics affecting the demand for different car-sharing
operational schemes. Transportation Research Interdisciplinary Perspectives, 14, 100616.

43
• Between the presence of children's status and economic status: the presence of children
may increase the desire to choose car-sharing (4 articles). However, it appears that the
children’s presence can reduce shared car demand in low-and-middle-income
households (1 article).
According to the above findings, the following policy implications and suggestions can be
formulated to expand the demand for different car-sharing schemes.
It ought to be noted that the rate of young members and people with above-average income
is higher among free-floating members. Also, males' adoption of this service is more elevated
than station-based service. Also, the rate of female members in free-floating services is higher
than in station-based services. Besides, as females seem more eager to opt for E-car-sharing
services, free-floating services can attract females by offering this type of car, especially in
Europe, where females are less attracted to car-sharing than females in North American
countries. Also, since an older age (55 years or older) of people living in households without
high income negatively affects the propensity to join a car-sharing scheme, free-floating
operators should target this group through specific actions.
It is also interesting to mention that although users of round-trip car-sharing seem less
educated than other car-sharing service users, car-sharing members may follow a more efficient
and sustainable lifestyle than the one-way shared car system members. For example, round-
trip service members have significantly fewer private cars than free-floating service members.
However, the rate of young members in the free-floating services is more elevated than in
station-based services. Since car-sharing with EVs has a special added attraction for young
couples with no private car, round-trip operators can offer this kind of service to attract younger
members. Furthermore, the probability of decreasing vehicle ownership by station-based
shared car members is higher than among free-floating car-sharing members. It may be because
members of free-floating shared car services do not necessarily see the service as a replacement
for their private car but as a supplement. Finally, it is important to note that car-sharing with
EVs has a special added attraction for young people who start a family and choose car-sharing
to complement their private car trips. Concerning developing e-car-sharing, some articles
identified the factors affecting the development or downturn of e-car-sharing services in the
entire e-car-sharing industry concerning stakeholders. (Turoń et al., 2020). Also, Turoń et al.
(2021) showed the main factors affecting the operation of the e-car-sharing market during the
COVID-19 and post-quarantine periods.
It should be noted that this study has some limitations. First, the results and claims related
to the effect of marital status and household size characteristics on car-sharing demand are
based on only a few articles. Therefore, more research needs to be done to increase the
robustness of the results, especially for free-floating and P2P services. In addition, more studies
should be done on the impacts of child presence and vehicle ownership characteristics on
demand for P2P services.
It is worth pointing out that although car-sharing has spread to the global markets, most
research on shared car systems has been investigated in China, the USA, Canada, and some

44
European countries. Hence, more studies need to be implemented in other countries, especially
in developing countries, to understand better the socio-demographic factors that affect car-
sharing demand according to the geographical area. For example, differences in education
levels between developed and underdeveloped countries may lead to different proportions of
car-sharing because there may be a relationship between education level and country. In
addition, other factors such as residence status (permanent residence or not, or tourist effect)
could be worth investigating to broaden the view.
Lastly, another research gap is the direction of causation between private car ownership
levels and car-sharing demand. There appears to be a complex two-way relationship between
car ownership and shared car demand. However, most studies have worked on the vehicle
ownership impacts on shared cars. Therefore, more research is required to work on vehicle
ownership as exogenous and endogenous variables to clarify the direction of causality and
better assess the shared car systems' sustainability.
Furthermore, the most remarkable trip-related characteristics that influence car-sharing use
are as follows:
• The higher the in-vehicle travel time and walking time to reach the nearest vehicle, the
less car-sharing usage is.
• Car-sharing is utilized chiefly for discretionary purposes.
The most important built environment and land use elements that affect the use of car-
sharing are as follows:
• Car-sharing users tend to live in dense urban areas with high public transportation
services.
The most considerable car-sharing characteristics that effects the use of car-sharing are as
follows:
• The higher the travel costs, the less car-sharing usage.
• The convenience of traveling by Free-floating car-sharing can increase usage.
The most significant attitudinal effects that impact car-sharing use are as follows.
• Environmental awareness is often seen among car-sharing users and shared electric
vehicles are preferred in this case.
• The private car status symbol can negatively affect car-sharing usage.
• The price effect is less pronounced for people with high psychological ownership; they
prefer private cars.
• Car-sharing users often use sustainable transport modes, such as public transportation
and non-motor modes.
• Technology dissemination seems to impact the acceptance of car-sharing positively.
• User satisfaction with the car-sharing service increases the likelihood that the person
will use the service later.

45
• The number of times users have utilized the service in recent months is proportional to
the current usage.
• Individuals' car-sharing choice behavior depends on their loved ones' car-sharing
choice behavior.
The influence of interactions between sub-factors on car-sharing use is as follows:
• Females are more likely to use car-sharing for short travel distances.
• Unemployed people or students in rural areas choose car-sharing as their mode of
transportation.
• Low psychological ownership may lead to a greater preference for car-sharing,
especially with low travel costs.
• The presence of children in low- and middle-income households can decrease the use
of car-sharing.
• As the travel distance increases, the user's willingness to use car-sharing rises in cold
weather.
• Travelers are more concerned with saving trip costs on shorter trips and saving travel
time on longer trips.
• For short trips, car-sharing is mostly used for non-commuting trips.
• For longer trips, users much prefer commuting trips.
• Electric car-sharing members are likely to retain their membership program mainly for
non-compulsory trips.
• Car-sharing is common in rural areas for non-commuting and longer trips.
Most studies have been carried out by considering only one or two main factors in the car-
sharing system. However, quantitative studies have considered several factors simultaneously.
Hence, more critical factors should be simultaneously considered in future research.
2.2 An overview of bike-sharing
The bike-sharing system and its benefit are explained in Chapter 1. This section offers an
overview of bike-sharing services to familiarize better with the important criteria and sub-
criteria that can influence bike-sharing use. In this regard, a brief history of bike-sharing,
integration of bike-sharing with other transport modes, bike, and its benefits, factors affecting
demand for bikes and summary, as well as factors affecting demand for bike-sharing, and its
summary are mentioned as follows.
2.2.1 A brief history of bike-sharing
As BSSs have proliferated, research on BSSs has emerged to ascertain the key attributes
leading to bike-sharing use. The BSS has a long history, and various BSS have popped up
worldwide (Si et al., 2019). According to Shaheen et al. (2010), there are four generations of
BSSs, the first of which was the “White Bikes” program, a BSS (unregulated) with the
installation of fifty unlocked and free bicycles dedicated to the public in Amsterdam, the
Netherlands in 1965. In this scheme, bikes are distinguished by a color painted in a light color,

46
and it was placed haphazardly and unlocked throughout the area so that everyone could use
them freely. However, the stolen bikes caused the program to fail (Eren and Uz, 2020).
In order to prevent bike theft, the “Coin-Deposit” system, the second bike-sharing
generation (Also known as Bycykel), was designed in Denmark in 1993. These bikes were
specially manufactured and distinguished by color or unique design (DeMaio, 2003). The bike
was locked and could be picked up and returned at designated bike stations throughout the city
with a coin deposit that incentivized people to return the bike to balance the BSS. The second-
generation systems were more expensive to operate than the first-generation ones. Both
generations of bike-sharing created more cycling opportunities. However, owing to the lack of
adequate support and reliable service, they could not induce people toward bike-sharing
transport mode (Bonnette, 2007).
In the 2000s, the third generation of BSPs, such as "Velo'v" in France and "Call a Bike" in
Munich, increased steadily over the decade. Also, the BSPs started to be established in other
countries such as China, the USA, and Brazil (Eren and Uz, 2020). The third generation, known
as “Station-Based Bike-Sharing” (SBBS), is an information-technology-based system that
introduced a more attractive BSS planning to increase people's encouragement to use bicycles
(Automated stations). This BSS is one of the intelligent transportation systems consisting of
innovative parking units, bike rental stations, and smart bikes (Raviv and Kolka, 2013). This
system employs kiosks or user interface technology, and bicycles are distinct by color, unique
design, or advertisements.
Furthermore, using innovative technology such as mobile phones, mag-stripe cards, or
smartcards, bikes can be picked up from the docking station and returned to each station
belonging to the same system. Also, this technology contributes to preventing bicycle theft
since members are required to provide identification, phone number, or bank card. Besides,
non-members usually have to pay a large deposit to ensure the bike's return. Therefore, the
integration of information technology has helped prevent bicycle theft. However, although the
third generation enticed more people to embrace the BSS, significant investments are required
to install adequate docking stations throughout the city.
The emergence of the fourth generation dockless bike, known as a “Free-Floating Bike-
Sharing” system, was due to requiring less investment. This free-floating bike system possesses
distinct bikes. This system is designed whereby people do not need to pick-up the bike from a
station or return it to the docking station. Instead, users can find the available bikes using an
embedded Global Positioning System (GPS). Given that the system does not require docking
stations and, therefore, does not require built-in infrastructure, the system has been rapidly
expanded globally (Shaheen et al., 2010; Shen et al., 2018). The smartphone application is
utilized in FFBSs, and the payment method is by scanning the Quick Response (QR) code or
by Near-Field Communication (NFC) (Shen et al., 2018).
Chen et al. (2020) compared the users’ attributes between the FFBS and SBBS in
Hangzhou, China. It was identified that the user structures for FFBS and SBBS are quite
similar, but the factors affecting the use frequency are different. SBBSs strength is providing a

47
low travel cost with appropriate quality, while FFBS is more convenient and flexible for users.
Therefore, the dockless design of the FFBS improves the users’ experience at the end of the
travel. Li et al. (2019) stated that the advent of FFBS has brought about essential changes in
urban cycling and the urban dwellers' transport mode choice. FFBS trips are suitable for
combination with bus and subway trips (Du et al., 2019). FFBS is ideal for connecting other
travel modes and the temporary travel demand (Li et al., 2019). The high efficiency and
flexibility of FFBS can integrate the BSS with public transport appropriately, leading to an
efficient alternative for first/last-mile travel (Chen et al., 2020).
Similarly, Li et al. (2018) stated that FFBS is an effective solution to the first/last mile
problem. In addition, the synergy of FFBS and public transport can increase BSS usage and
enhance the benefits of both modes (Shen et al., 2018). Finally, Shaheen et al. (2012) pointed
out that the fourth-generation targets efficiency, quality, and sustainability.
On the contrary, Sun (2018) noted that dockless BSSs yield negative consequences such
as abatement of public space and bike-share vandalism. Also, these systems are not a substitute
for private vehicles. Besides, oversupply has led to graveyards of bikes and deep concerns
about maintenance, quality control, and management of these systems. Li et al. (2018) pointed
out that the lack of policy for FFBS and delays in fixing bike defects are the major hurdles
standing in the way of increasing FFBS usage in Jiangsu, China. Du and Cheng (2018) noted
that if FFBS malfunctions are not addressed promptly, it can impede utilization or reduce the
usage rate. Also, bicycle availability and easy finding are important factors in increasing FFBS
demand.
2.2.2 Integration of bike-sharing with other transport modes
Cities across the globe are embracing BSSs, and people tend to integrate the bicycle-sharing
journey into their daily travels (Schoner and Levinson, 2013; Mateo-Babiano et al., 2016).
Because the BSS integrates cycling into the transportation system, it increases the mobility
option providing a more convenient and attractive transport mode for users. Therefore, one
feature that can affect BSS's success is its integration with effective public transport
interchanges (Jennings, 2011; Bagloee et al., 2016). The purported benefits of BSS promote
inner-city public transport options (Vogel and Mattfeld, 2011). Travel time is reduced when
the BSS is well-integrated with the public transport network (McBain and Caulfield, 2018). A
study carried out in Helsinki, Finland, by Jäppinen et al. (2013) determined that the use of BSP
decreased the travel time of public transport by more than 10%, meaning about 6 min. Hence,
the BSP strengthens public transport, enhances connections, and improves sustainable daily
mobility (Shaheen, 2012; Jäppinen et al., 2013). The "Call-A-Bike" in Germany and the
"Vélo'v," launched in May 2005 in Lyon, France cities, are examples of BSPs, deployed at
public transport stops (Borgnat et al., 2011; Buehler and Pucher, 2011).
The BSSs target daily mobility and people choose the BSS on an as-needed basis (Hyland
et al., 2018). Ma et al. (2019) mentioned that two-thirds of car drivers were willing to use Free-
Floating Bike-Sharing (FFBS) for short-distance trips (within 2 km) in Nanjing, China. In
addition, Perceived health, perceived ease of use, and perceived usefulness positively affect

48
individuals' attitudes toward FFBS. Also, it was found that individuals' attitudes toward FFBS
positively correlate with their willingness to shift. Also, BSS is utilized for access and egress
to transit during peak hours (Noland et al., 2019). Levy et al. (2019) noted that buses could
complement bikes for shorter trips, most concentrated in the city center. Besides, BSSs seem
to be substituting buses for longer trips, most of which are focused on links dedicated to bike
lanes. Shaheen et al. (2011) determined that the BSS acted as a complementary competitor to
public transportation, and also, BSS seems to decrease car trips. Fishman et al. (2014)
determined that due to the use of BSS, the alleviation in motor vehicle use was roughly 90,000
km per year in Minneapolis and Melbourne. Ricci (2015) noted that bike-sharing trips are
predominantly utilized instead of public transportation and walking trips and are not a potential
alternative to car trips. Also, some studies found that the BSS mainly substitutes for walking
rather than public transportation and cars (Murphy and Usher, 2015; Zhang, 2017). Du et al.
(2019) mentioned that FFBS attracted users whose main transport modes are private bikes
(15%), walking (39%), and conventional buses (14%) in Nanjing, China.
2.2.3 Bike and its benefits
Before reviewing the factors influencing bike-sharing demand, an overview of the factors
influencing bicycle selection can be contributory. Increased dependence on private vehicles
imposes high social, economic, and environmental costs, which are likely to surge in traffic,
raise energy consumption, and, to owe to the increased vehicle source emissions, degrade air
quality (Litman and Laube, 2002; Saelens et al., 2003; Sener et al., 2009). Improving cycling
to school can lead to increased healthy travel behaviors (Forsyth and Oakes, 2015). This is
likely to be maintained in adulthood, helping the next generation develop greener travel
behavior. Thus, urban planning and public health officials have been steadfast in persuading
people to use active transportation modes in recent decades (Krizek et al., 2007).
According to reports, although nowadays people choose motorized vehicles for short trips,
the "future belongs to walking and cycling" (Davis et al., 2012). Cycling is established as one
of the best options among urban mobility alternatives since its facilities do not require much
space; it is environmentally friendly and positively affects health, which is an important issue.
Especially the physically inactive lifestyle is a significant challenge to public health (Sallis et
al., 2004). As transportation is a routine in which we all engage, cycling has excellent potential
to surge the level of daily physical activity (Strong et al., 2005). Besides, in urban areas of
developed countries, the travel time of half of the trips can be less than 20 minutes by bike
(Kamargianni, 2015). Thus, the growing presence of cycling can increase its role alongside
other aims at promoting sustainable transport. For instance, it helps alleviate social,
environmental, energy, and traffic congestion and concerns about the high rate of car use, and
it could provide substantial health benefits (Wardman et al., 2007). People should change their
travel behavior to lessen the deleterious effects of the private vehicle and achieve closely
aligned objectives, including enhanced livability, raised physical activity, reduced traffic
congestion, and reduced levels of air pollution. For instance, choosing a bike instead of a
private vehicle can assist in obtaining these aims, such as decreasing vehicle-generated air
pollution.

49
2.2.4 Factors affecting demand for bike
A wide-ranging set of factors influencing cycling behaviors has been studied recently. Exited
literature has identified several factors that can influence bike choice. These factors can be
categorized into five categories: 1. socio-demographic characteristics, 2. trip-related
characteristics, 3. built environment and land use, 4. bike characteristics, and 5. natural
environmental conditions.
2.2.4.1 Socio-demographic characteristics
The socio-demographic characteristics, including gender, age, education level and awareness,
occupation and economic status, and ownership status, affect bike use.
2.2.4.1.1 Gender
Some studies found that males' cycling trips usually surpass females (Shafizadeh and Niemeier,
1997; Parkin et al., 2008; Baker, 2009). The influence of the cyclist's gender on mode choice
behavior is due to the gender differences in risk aversion (Garrard et al., 2008). Also, females'
perceptions of the feasibility of alternative transportation modes differ from that of males. For
instance, to use a bike, the importance of being proximate to bicycle trails and paths is more
for females than males (Akar et al., 2013). Female commuters are more inclined to choose the
car for home-based school (HBSc) trips rather than walking and cycling, consistent with some
surveys (Mota et al., 2007; Larsen et al., 2009). In girls' HBSc trips, street connectivity
positively correlates with active commuting to school (Mota et al., 2007). Generally, Females
prefer private motorized vehicles over active transport (Clifton, 2003; Timperio et al., 2006).
Further, it is important to note that females are willing to cycle on routes with maximum
separation along heavily traveled roads. When considering the existing cycle paths network,
females value adequate and safe paths more than males (Kamargianni, 2015). Hence, providing
bicycle paths and lanes obtaining a high degree of separation from motor traffic may be
significant for raising bike commute rates amongst females (Garrard et al., 2008). As
mentioned, feeling safe is positively associated with cycling choices (Akar et al., 2013).
2.2.4.1.2 Age
According to Shafizadeh and Niemeier (1997), younger commuters may be less willing to
make longer commutes than the elders. Focusing on the age factor impacting the utility of
bicycles, it is determined that youngsters consciously avoid private motorized vehicles (Davis
et al., 2012; Axhausen, 2013). They have selected this lifestyle regardless of their income status
(Kamargianni, 2015).
2.2.4.1.3 Education level and awareness
Ortuzar et al. (2000) explained that there is a lack of proper understanding of cycling in some
areas. For instance, in Chile, there was public ridicule of riders on network television stations.
In a household, the parent's attitudes toward cycling and obtaining a high level of education
(bachelor's degree) by the mother significantly affect the teenager's desire to cycle. In General,
people with college educations are more likely to choose cycling (Barnes and Krizek, 2005;
Xing et al., 2010). Hence, people's culture and education influence bicycle use (Kamargianni
and Polydoropoulou, 2013). Generally, traffic education and training for drivers and cyclists,
creating enthusiasm to cycle, and broad public transport which supports cycling have raised

50
the cycling levels (Pucher and Buehler, 2008). Also, the availability of school courses for safety
skills on how to walk and cycle safely can grow cycling rates (Kamargianni, 2015).
2.2.4.1.4 Occupation and economic status
Some researchers mentioned that high income reduces the utility of active transport (Jara-Díaz
and Videla, 1989; Sallis et al., 2004). Xing et al. (2010) argued that people with higher incomes,
compared to individuals with lower incomes, would choose faster modes as they attach higher
values to their time. Hence, the higher pocket money diminutions the utility of cycling to school
(Kamargianni and Polydoropoulou, 2013). Nevertheless, in rural areas, with increasing pocket
money, teenagers still prefer to ride a bike (Kamargianni, 2015).
2.2.4.1.5 Ownership status
Car ownership hurts the cycling demand. However, imposing a tax on car ownership and
parking creates unpleasant and expensive driving in central cities, leading to higher cycling
rates (Parkin et al., 2008).
2.2.4.2 Trip-related characteristics
Trip-related characteristics include travel time, trip purpose, and travel distance that impact
bicycle use.
2.2.4.2.1 Travel time
People are generally sensitive to increasing trip length, represented as higher journey time,
especially for non-motorized modes (Akar et al., 2013). Similarly, increasing the travel time of
bicycles to school may result in adolescents refusing to choose bikes (Kamargianni and
Polydoropoulou, 2013; Kamargiani, 2015). Usually, travelers younger than 24 and females pay
more attention to travel time while choosing a bike (Krizek et al., 2005; Garrard et al., 2008;
Dell'Olio et al., 2014). Whalen et al. (2013) mentioned that using a bicycle while traveling for
less than 10 minutes can be overlooked because of the overuse of other modes. However, if the
travel time is increased to 10 minutes, it may positively affect cycling and increase the bike's
share to more than 14%. Therefore, there should be a travel time range in which the cycle use
rate increases and begins to decrease for travel time beyond the range.
2.2.4.2.2 Travel distance
One of the considerable factors affecting bike use is travel distance. Long distance negatively
impacts children's active movement (Timperio et al., 2006). Experienced cyclists can cycle the
bike for a longer distance than other kinds of cyclists. Xing et al. (2010) showed that perceived
trip distance influences cycling choice. Ortúzar et al. (2000) examined the fundamental factors
conditioning use of bicycles in Santiago. It was determined that the trip length is one of the
most influential factors in selecting cycling as an alternative mode of transport. Furthermore,
it was found that although short trips are the most important market for bikes, a reduction in
trip length and adequate incentive for metro and suburban railway station transfers can increase
the level of cycling in a large city.
2.2.4.2.3 Trip purpose
Xing et al. (2010) focused on the trip purpose effect on cycle choice and mentioned that cycling
usually is used for recreational trips.

51
2.2.4.3 Bike characteristics
Bike characteristics such as travel costs can significantly affect bike usage.
2.2.4.3.1 Travel cost
Travel costs can significantly affect adolescents and adults on their mode choice behavior
(Kamargianni and Polydoropoulou, 2013). Wardman et al. (2007) found that a £2 per day for
employees who cycle to work is highly effective and not far from doubling the amount of
cycling and a 5.4% reduction in car demand. A £5 daily payment can also decrease car demand
by 23.6%.
2.2.4.4 Built environment and land use
Land use, accessibility, infrastructure, trip end, and en-route facilities are substantial elements
impacting bike usage.
2.2.4.4.1 Land use
It is important to note that land use is essential in cycling choices. For example, the employment
densities at destinations, compared to the residential densities at origins, have more impact on
the mode choice for home-based work (HBW) trips (Rodrıguez and Joo, 2004). Also, the built
environment is correlated with the children's active commuting to school (Kerr et al., 2006).
Larsen et al. (2009) reported that the more land use mix and the presence of street trees, the
more use of active transport modes in HBSc trips. Also, Winters et al. (2011) determined that
scenic bike routes (aesthetically pleasing locations), traffic-calmed streets, trails segregated
from motorized traffic, and away from traffic noise and pollution can be an important incentive
for cyclists. Whereas streets with high-speed traffic and the risks from motorists are the top
deterrent factors.
2.2.4.4.2 Accessibility
Accessibility is also a significant factor, as nearness to trails and the presence of agglomerations
of hospitals, fast-food restaurants, offices, and clinics are influential environmental factors on
cycling choices (Maurer et al., 2012).
2.2.4.4.3 Infrastructure, Trip End, and En-Route Facilities
Some studies investigated the importance of providing ample cycling facilities, including
parking space availability, off-road and in-traffic facilities, bike paths, and lanes (Bowman et
al., 1994; Nelson and Allen, 1997; Ortúzar et al., 2000; Dill and Carr, 2003; Krizek et al., 2007;
Tilahun et al., 2007; Garrard et al., 2008; Pucher and Buehler, 2008; Krizek et al., 2009; Dill,
2009; Larsen and El-Geneidy, 2011; Buehler and Pucher, 2012; Kamargianni and
Polydoropoulou, 2013; Bhat, 2015). The provision of infrastructure would infer the
construction of more cycle paths across the city to elevate the convenience and safety of riders.
Installing a network of bicycle rental stations can boost the accessibility of potential bike users
who do not have bicycles. Hence, doing so would likely have significant implications for
encouraging increased cycling levels (Nelson and Allen, 1997; Dill and Carr, 2003; Wardman
et al., 2007; Hunt and Abraham, 2007; Dill, 2009; Handy et al., 2010; Winters et al., 2011;
Buehler and Pucher, 2012).
Wardman et al. (2007) created a comprehensive model to predict future trends in urban
commuting shares over time and the effects of different measures to increase the willingness

52
to cycle to work. The results indicated that the en-route cycling facilities, utterly segregated
cycleways, have the highest effect on cycling choice. However, the results showed only a 55%
increase in cycling and a slight decrease in car commuting. Pucher and Buehler's (2008)
research explains why cycling has become a relatively convenient, safe, and practical way to
travel around cities in Dutch, Danish, and German cities where cycling is a way of life. It is
clarified that the most important factor which has enticed people to cycle is generating separate
cycling facilities along intersections and busy roads, coupled with the traffic calming of most
residential areas. In addition, the broad cycling rights of way, adequate bike parking, and a
fully integrated bike system with public transport have affected the bike use rates.
Dell'Olio et al. (2014) recognized the potential of cycling as a sustainable mode of transport
in Santander, a medium-sized city with steep streets and relatively inclement weather in Spain.
The results revealed that an extensive network of public and private bike docking stations is
significantly more valuable than a network of cycle paths, which can ensure comfortable and
safe cycling in the city.
Winters et al. (2011) indicated that the factors associated with the built environment for
cycling, such as separation from motor vehicles, pleasant route conditions, and ease of cycling,
significantly affect bike choice. In addition, the presence of bicycle signage and traffic signals
leads to higher levels of bicycle commuting (Winters et al., 2010). They compared the modal
split gained as a function of the existence of cycle paths with that of docking stations less than
400 m away. They determined that the ease of bicycle parking is more important than traveling
safely and comfortably for bicycle commuters (Dell' Olio et al., 2014). Also, with the growing
coverage of cycle lanes (i.e., cycle paths painted on the pavement but not segregated from the
traffic) and cycleways between home and school, teens are more likely to choose cycling for
the HBSc trips (Kamargianni and Polydoropoulou, 2013). Contrastly, poor access to lights or
crossings and busy intersections between the home and school adversely affect the children's
active commuting (Timperio et al., 2006). Beginner cyclists appreciate the presence of bike
lanes 1.6 times more than experienced cyclists (Motoaki and Daziano, 2015) because a
separated path or striped lane can augment a cyclist’s perception of safety (Dill and Carr, 2003).
According to Hunt and Abraham (2007), the time spent cycling in mixed traffic is more
onerous than time spent cycling on bike paths. Certainly, cycleways are safer than cycle lanes,
and cycleways can increase the tendency to cycle more (Ortúzar et al., 2000). Hence, streets
with separate paths, bicycle lanes, bicycle boulevards, and well-connected neighborhood
streets can attract more adult cyclists (Dill, 2009). Furthermore, in rural areas, the coverage of
cycleways is the most influential factor in choosing a bicycle (Kamargianni, 2015). In general,
cities that possess cycling facilities in the right places witnesses a higher level of bike
commuting and also a proper design that considers the type of the city could increase the
cycling propensity (Krizek and Roland, 2005; Tilahun et al., 2007; Krizek et al., 2009; Winters
et al., 2010; Winters et al., 2011; Flugel et al., 2015).
Furthermore, providing secure parking compared to showers attracts more people to cycle
(Hunt and Abraham, 2007). The presence of bicycle parking lots in the schoolyard favors
choosing a bicycle for adult students because they possess a place to park and lock their

53
bicycles during school hours (Kamargianni, 2015). Hence, given the availability of safe and
convenient infrastructure and the right built environment, people persuade to opt for cycling
for their short trips (Kamargianni, 2015).
According to Wardman et al. (2007), the individuals who may have the willingness to
select cycling as a mode of transport are not necessarily a homogenous group. Hence, providing
packages of measures that include a range of motivations for cycling promotion is the best
approach to enhance the propensity to cycle. Therefore, the most effective policy to increase
the demand for cycling to work is to combine the amelioration of en-route facilities, the daily
payment to cycle to work, and comprehensive trip-end facilities (provision of showers and
indoor parking at the workplace). It would also have a considerable effect on decreasing the
level of car use. In addition, Handy et al. (2010) determined that employing a comprehensive
package of strategies targeting the factors of the individual, physical environment, and social
environment has synergistic effects that are the best approach to raising cycling levels. Finally,
it is important to note that adopting some practical policies, such as introducing a city center
congestion charge for private cars, could heighten the economic value of the activity. Enacting
this policy is likely to amend the negative situation and, by changing people’s attitudes toward
cycling, leads to inducing them to use bicycles in such areas.
2.2.4.5 Natural environmental conditions
The hilliness, weather conditions, temperature, humidity level, and air pollution factors are
natural environmental conditions that influence bike use.
2.2.4.5.1 Hilliness
The road with a lower gradient attracts more cyclists (Waldman, 1977; Rietveld and Daniel,
2004; Rodrıguez and Joo, 2004; Timperio et al., 2006; Parkin et al., 2008; Winters et al., 2010).
2.2.4.5.2 Weather condition
Weather condition is one of the most dominant factors in using a bike (Parkin et al., 2008;
Kamargianni and Polydoropoulou, 2013; Dell'Olio et al., 2014; Wang, 2015). A study by
Kamargianni (2015) investigated the factors impacting bike use for HBSc trips in different
areas. It was determined that inclement weather conditions have the most significant impact on
bicycle selection in urban areas. Dell' Olio et al. (2014) indicated a significant difference in the
modal split between good and bad weather. When the weather is unfavorable, the most
privileged mode for potential bicycle users is the private car. Cycling is the most attractive
mode of transport when the weather is favorable. Winters et al. (2011) mentioned that snow
and ice could decrease bike use. Nankervis (1999) studied the seasonal and weather-related
variation patterns that indicated a decrease in bike use in winter and under inclement weather
conditions. Ortúzar et al. (2000) demonstrated that weather (sunny days) is the factor that
affects the utility of bicycles the most in all areas except the rural areas, where the most
important factor is the percentage of cycleway coverage on the route between home and school.
Even when the average temperature is minus 12 centigrade, teenagers use the bike to transport
to school in rural areas. Commonly, even when the weather is sunny, females from all areas do
not prefer to cycle. Motoaki and Daziano (2015) found that the adverse effect of rain and snow

54
on less-skilled cyclists is 2.5 and 4 times higher, respectively, compared to cyclists with higher
skills.
2.2.4.5.3 Temperature
One factor that can significantly impact the choice of cycling is the temperature (Parkin et al.,
2008; Kamargianni and Polydoropoulou, 2013; Wang, 2015). According to Saneinejad et al.
(2012), in Toronto, the levels of bike use are sensitive to wind speed and temperature only in
conditions below 15 centigrade. In addition, the adverse effects of cold temperatures on the use
of bicycles are higher for young people than for the elderly. Also, females' level of bike use is
about 1.5 times more likely to be influenced by cold temperatures than males.
2.2.4.5.4 Humidity level
Cycling is generally incompatible with high humidity (Zahran et al., 2008).
2.2.4.5.5 Air pollution
Cycling is incompatible with high air pollution (Zahran et al., 2008).
2.2.5 Summary
As mentioned above, the natural environmental conditions, including the weather condition,
temperature, humidity level, air pollution, and hilliness, are important factors in cycling choice.
Moreover, the built environment and land use are significant factors in choosing a bike. For
instance, land use, accessibility, and the provision of infrastructures, such as encompassing the
coverage of cycle lanes and cycleways, the availability of bicycle parking lots, safety, and the
introduction of public bicycle docking stations, are important factors. In addition, socio-
demographic characteristics comprising gender, age, occupation and economic status,
education level and awareness, and ownership status are influential factors in choosing a bike.
In addition, bicycle characteristics such as travel costs can affect bicycle use. Also, trip-related
characteristics such as trip purpose, distance, and travel time could be another object that must
be addressed. To sum up, the factors affecting bike choice are indicated in Table 23.
Table 23: The effect of different factors on bicycle use.
Factors
Sub-factors
Positive impact
Negative
impact
References (bike docking
station is not studied)
References
(bike docking
station is
studied)
Natural
environmental
conditions
Weather
condition
Favorable weather
(sunny)
Unfavorable
weather
(windy, rainy,
and snowy)
Nankervis, 1999; Ortúzar et al.,
2000; Parkin et al., 2008;
Winters et al., 2011; Saneinejad
et al., 2012; Kamargianni and
Polydoropoulou, 2013;
Kamargianni, 2015; Wang,
2015; Motoaki and Daziano,
2015
Dell'Olio et
al., 2014
Temperature
Cold
Ortúzar et al., 2000; Saneinejad
et al., 2012
Humidity level
High
Zahran et al., 2008
Air pollution
High
Zahran et al., 2008
Winters et al.,
2011
Hilliness
Low gradients
High Gradients
Waldman, 1977; Rietveld and
Daniel, 2004; Rodrıguez and
Joo, 2004; Timperio et al., 2006;
Parkin et al., 2008; Winters et
al., 2010

55
Factors
Sub-factors
Positive impact
Negative
impact
References (bike docking
station is not studied)
References
(bike docking
station is
studied)
Built
environment and
land use
Land use
More land use mix,
trees through the
route, higher
employment, and
population density
Rodrıguez and Joo, 2004;
Larsen et al., 2009
Accessibility
Near to important
places (trails, fast-
food restaurants,
hospitals, clinics, and
offices)
Poor access to
lights or
crossings,
Timperio et al., 2006; Maurer et
al., 2012
Infrastructure,
trip end, and En-
route facilities
Infrastructure, trip
end, and En-route
facilities (cycleways,
cycle lanes, showers,
bike parking, bicycle
sign, age, traffic
signals, bike docking
station, safety, and
comfort)
High-speed
traffic
Bowman et al., 1994; Nelson
and Allen, 1997; Ortúzar et al.,
2000; Dill and Carr, 2003; Kerr
et al., 2006; Timperio et al.,
2006; Krizek et al., 2007; Hunt
and Abraham, 2007; Wardman
et al., 2007; Tilahun et al., 2007;
Garrard et al., 2008; Pucher and
Buehler, 2008; Krizek et al.,
2009; Dill, 2009; Winters et al.,
2010; Handy et al., 2010;
Buehler and Pucher, 2012;
Kamargianni and
Polydoropoulou, 2013;
Kamargianni, 2015; Bhat, 2015
Winters et al.,
2011;
Dell'Olio et
al., 2014
Trip-related
characteristics
Trip purpose
Recreational trips
Xing et al., 2010; Larsen and El-
Geneidy, 2011
Trip distance
Long-distance
Timperio et al., 2006; Xing et
al., 2010; Kamargianni and
Polydoropoulou, 2013
Travel time
Short trips (like 10
minutes)
Long trips
Ortúzar et al., 2000; Krizek et
al., 2005; Garrard et al., 2008;
Kamargianni and
Polydoropoulou, 2013; Akar et
al., 2013; Whalen et al., 2013;
Kamargianni, 2015
Dell'Olio et
al., 2014
Bike
characteristics
Travel cost
Daily payment to
employees who cycle
to work, charge
private cars
High public
bike rental rates
Wardman et al., 2007; Pucher
and Buehler, 2008; Parkin et al.,
2008
Dell'Olio et
al., 2014
Socio-
demographic
characteristics
Age
Young
Old
Shafizadeh and Niemeier, 1997;
Davis et al., 2012; Axhausen,
2013; Kamargianni, 2015
Gender
Male
Female
Shafizadeh and Niemeier, 1997;
Clifton, 2003; Timperio et al.,
2006; Mota et al., 2007; Garrard
et al., 2008; Parkin et al., 2008;
Baker, 2009; Larsen et al., 2009;
Akar et al., 2013; Kamargianni,
2015
Occupation and
economic status
Lower-income
Higher-income
Jara-Díaz and Videla, 1989;
Sallis et al., 2004; Xing et al.,
2010; Kamargianni and
Polydoropoulou, 2013;
Kamargianni, 2015
Ownership status
Car ownership
Parkin et al., 2008;
Education level
and awareness
College education
received traffic
Education (school
course on safety)
Ortuzar et al., 2000; Barnes and
Krizek, 2005; Wardman et al.,
2007; Pucher and Buehler, 2008;
Xing et al., 2010; Kamargianni
and Polydoropoulou, 2013;
Kamargianni, 2015
Winters et al.,
2011
2.2.6 Factors affecting demand for bike-sharing
BSPs have spread swiftly throughout the world in recent decades (Tang et al., 2011). The
benefits of the BSS can lead many people to choose it as a transport alternative mode that

56
makes the BSS worthwhile for investment. Accurate estimation of bike-sharing demand is an
important factor in the success of BSSs (Jennings, 2011). Therefore, urban planning agencies
should predict bike-sharing demand to make investment decisions (Skov-Petersen et al., 2017).
Also, identifying the factors influencing bike-sharing ridership is essential for policymaking
(Duran-Rodas et al., 2019). Hence, it is necessary to examine the elements substantially
affecting the levels of bicycle use. In order to achieve this aim, it is significant to take into
account various elements such as social, individual, and other environmental influences to
approach this field of study from a holistic perspective. The literature on the BSS has shed light
on the key factors contributing to bike-sharing demand that can help assess the performance of
BSPs comprehensively and would pave the way for building a complete and articulated picture
of BSS's different aspects. Factors influencing the demand for bike-sharing can be classified
into five characteristics: the socio-demographic, trip-related, bike-sharing, built environment
and land use, and natural environmental conditions.
2.2.6.1 Socio-demographic characteristics
Examining the socio-demographic characteristics, including age, income, gender, residence
status, and education level, is imperative to building a deeper understanding of the user profile
of BSPs and boosting the users' loyalty and retaining them (Rixey, 2013; Li et al., 2019). Some
socio-economic features, such as gender, ownership status, and employment, can affect users'
willingness to use BSS more than travel restrictions strategies (Feng and Li, 2016).
2.2.6.1.1 Gender
Gender is also a considerable factor influencing BSS usage (Nikitas, 2018). A recent review of
the scientific literature has concluded that males are more avid users of BSSs compared to
females (Ogilvie and Goodman, 2012; Vogel et al., 2014; Ricci, 2015; Fishman, 2016; Raux
et al., 2017; Du and Cheng, 2018). Also, Chen et al. (2020) noted that the proportion of male
users is more than females for both SBBS and the FFBS in Hangzhou, China. The gender effect
on BSP can be exemplified by the percentage of male users in Melbourne, 76.6%, in Brisbane,
59.8% (Buehler and Hamre, 2014), and the proportion in Montreal, 58% (Bachand-Marleau et
al., 2012). Also, according to the results of the surveys conducted by Zanotto (2014) in
Vancouver, Canada, 52.8% of BSS members were males.
Furthermore, females accounted for only 21% of Chicago’s Divvy BSS members in
Chicago USA (Faghih-Imani and Eluru, 2015). Besides, Goodman and Cheshire (2014)
showed that less than 20% of London BSP members are females. Also, Wang and Akar (2019)
reported that more than two-thirds of the bike share trips were made by males in New York
City, USA. Also, female users are more sensitive to traffic conditions and make fewer
commuting trips. Further, the number of subway entrances and bus stops around bike-share
stations negatively impacts females' use of bicycles. In addition, males may link bike-share
trips to public transit services more than females. It is essential to state that, based on the Li et
al. (2019) study, when the travel distance is between 4 km and 8 km, females are more likely
than males to choose PBS for their travels.
2.2.6.1.2 Age
Age is a significant element in using BSSs (Raux et al., 2017; Nikitas, 2018). Older people
tend to select PBS (Li et al., 2019). In contrast, young people prefer to choose the FFBS for

57
their travels (Du and Cheng, 2018; Li et al., 2019). Vogel et al. (2014) revealed that the 18-49
age group accounts for about 80% of the total number of active subscribers and users in Vélo’v,
Lyon. In general, young people are more likely to be involved in cycling than elderlies (Fuller
et al., 2011; Ricci, 2015; Eren and Uz, 2020). Also, Chen et al. (2020) mentioned that in both
the SBBS and the FFBS, most users are younger than 35 years old in Hangzhou, China.
Besides, Wing et al. (2018) pointed out that the BSP is mainly used by the 28 to 37 age cohort
in Manhattan, New York. Besides, Fishman et al. (2015) noted that 16.9% of BSS users were
between 30 and 34 years old in Melbourne, Australia. Also, Zanotto (2014) mentioned that
67.5% of BSS members were between 16-54 years of age in Vancouver, Canada.
Correspondingly, individuals aged 18-34 are 3.3 times more likely than other age groups to be
members of the BSP in Australia.
2.2.6.1.3 Education level
Furthermore, education level is one of the influential factors in using bicycle-sharing systems
(Fuller et al., 2011). BSS users are probably highly educated (Ricci, 2015; Du and Cheng,
2018; Li et al., 2019). For example, in the “Capital Bike-Share” program in Washington, DC,
95% of users have a four-year college degree, 56% of whom possess an advanced degree
(Bachand-Marleau et al., 2012). Besides, Fishman et al. (2015) mentioned that 81% possessed
a bachelor's degree or higher education. Also, Zanotto (2014) stated that 65.6% of BSS
members had post-secondary education in Vancouver, Canada. In a study by Cheng et al.
(2020) in Hangzhou, China, it was found that FFBS and SBBS users had at least a bachelor’s
degree. Also, for the SBBS, possessing a graduate-level degree was associated with higher use
of SBBS, but not for the FFBS.
2.2.6.1.4 Occupation and economic status
Income is another factor impacting bike-sharing usage (Maurer, 2011). Affluent people are
inclined to choose PBS (Fishman et al., 2015; Ricci, 2015; Murphy and Usher, 2015; Raux et
al., 2017). Also, Li et al. (2018) noted that people with high incomes were more likely to use
FFBS in Jiangsu, China. Besides, Rixey (2013) mentioned that mid-income positively relates
to BSS usage. However, some barriers exist for low-income groups, such as providing credit
card information or accessing the internet to receive a long-term bike-sharing rental card
(Murphy and Usher, 2015). The results of an online survey set out by Fishman et al. (2015) in
Melbourne, Australia, stated that 43% of the BSS users received an annual salary of 104,000
dollars or more. Also, according to the results of the surveys conducted by Zanotto (2014) in
Vancouver, Canada, 72% of BSS members were employed, and 57.9% had an annual income
of 50,000 dollars or higher.
2.2.6.1.5 Ownership status
Vehicle ownership is another factor in studying bike-sharing usage rates (Fishman et al., 2015).
Shaheen and Guzman (2011) stated that the BSP members (22%) had higher car ownership
rates than non-members (11%) in Hangzhou, China, in 2010. Fishman et al. (2015) found that
76.6% of the BSS users owned a car in Melbourne, Australia. Hence, car ownership does not
appear to decrease the likelihood of bike-sharing usage.
On the other hand, Chen et al. (2020) stated that most SBBS and FFBS users did not own a car
or e-bike in Hangzhou, China.

58
2.2.6.1.6 Residence status
Interestingly, the residence status of individuals affects the use of BSS. There is a difference in
user confidence in the BSS; people with permanent residency are more likely to use PBS, while
people without permanent residency prefer FFBS (Li et al., 2019). Du et al. (2019) reported
that the residents without registered permanent residence use FFBS, residents with registered
permanent residences use fewer FFBS systems, and most (64.68%) own private cars in
Nanjing, China.
2.2.6.2 Trip-related characteristics
Trip-related characteristics contain travel time, departure time, travel distance, and trip purpose
impact using bike-sharing.
2.2.6.2.1 Travel time
According to Buehler and Hamre (2014), because of the travel time (73% of users) savings in
Washington, DC, many Capital Bikeshare (CaBi) riders tend to choose the bike-sharing
transport mode. The Results of a study by Mateo-Babiano et al. (2016) revealed that the free
initial period under the CityCycle program in Brisbane, Australia, has persuaded most users to
choose short-term trips for not incurring any charges other than membership. Similarly, Ahillen
et al. (2016) stated that the PBS program was utilized on short trips. Jensen et al. (2010)
characterized the speed and paths of bike-sharing usage in Lyon, France. It was stated that
using BSP for short trips with high-speed travel is prevalent. It was found that with almost no
traffic lights or car impedance, the average speed of bike-sharing reached 14.5 km /h in the
early morning of the week. Besides, when there were shortcuts to bicycle travel, most bicycle
trips were shorter than car trips. It was also presented that when you are less in a hurry to reach
a destination, such as traveling on weekend afternoons, the average travel speed is reduced to
10 kilometers per hour.
2.2.6.2.2 Departure time
The departure time is an important feature to consider. Generally, there are morning and
evening peak-hour demands (Kaltenbrunner et al., 2010; Ahillen et al., 2016). Ahillen et al.
(2016) found that the PBS program's demand for bikes rose in the morning and afternoon rush
hours. Similarly, Ji et al. (2020) found that the departure time in the morning rush hours (7 am-
9 am) and the afternoon peak hours (5 pm-7 pm) is positively correlated to both the SBBS and
the FFBS usage on workdays. Zhang and Mi (2018) found that the peak hours are between 7
and 8 in the morning and between 5 and 6 in the evening.
Li et al. (2019) mentioned that the demand for bike-sharing is low in the afternoons in
Beijing, China. However, Du and Cheng (2018) reported that the evening peak was more
significant than the morning peak. Also, Faghih-Imani et al. (2014) indicated that the bike-
sharing flow is higher in the evening than in the morning in Montreal, Canada. In this regard,
Reiss and Bogenberger (2016) segmented the operating area into 40 zones in Munich,
Germany. It turned out that the morning demand for bike-sharing rent was more than in the
afternoon and evening at the edge of the operating area.
Conversely, in zones near the city center, the demand for bike-sharing was higher in the
evening than in the morning. Also, in a study by Froehlich et al. (2008), usage patterns of the
bike-sharing scheme in Barcelona, Spain, showed a rise in bike-sharing use from residential

59
areas to commercial areas at 7 am on weekdays. Furthermore, the demand for BSS from
commercial to residential areas increased after working hours.
Kim et al. (2012) identified the difference between weekday and weekend travel behavior
in demand for BSSs in Goyang, South Korea. The latter possesses twice the amount of bike-
sharing demand compared to the former. Also, the “CityCycle” scheme, the most extensive
PBS program in Australia, is used chiefly on weekends for leisure in Brisbane (Mateo-Babiano
et al., 2016).
In contrast, Corcoran et al. (2014) presented a general system-wide decrease in the number
of travels taking place on weekends in Brisbane, Australia. Besides, Faghih-Imani et al. (2014)
reported that the bike-sharing demand decreased over the weekend in Montreal, Canada. In
addition, Heaney et al. (2019) reported that people were likelier to choose BSS on weekdays
in New York City, USA. Also, Lin et al. (2020) indicated that daily bike-sharing use reduces
by roughly 51.5% on public holidays or weekends compared to the workday in Beijing, China.
O’Brien et al. (2014) stated that there is no noticeable difference in the passenger flow of
the BSS between weekends and weekdays in Washington, DC. Similarly, Kim et al. (2018)
showed no remarkable difference in the number of bike-sharing rentals on weekdays and
weekends, but on the weekend morning, the number of trips was reduced. In this regard, Kutela
and Kidando (2017) mentioned that compared to evening peak hours (4 pm to 6 pm) and
weekends, the morning peak hours and weekdays are accompanied by an increase in the
likelihood of the Bikes Idle Duration (BID), respectively. Finally, it is worth noting that using
intelligent public transport cards for bicycle rental can persuade users to use the BSS at night.
2.2.6.2.3 Trip purpose
A precise understanding of the trip purpose factor can aid in better comprehending the travel
demand and the distribution of rental stations, which is essential information for planning the
BSS (Li, 2019). Fishman (2016) noted that BSS annual members' most common trip purpose
is commuting. Besides, Chen et al. (2020) found that the top three travel purposes for the
SBBS and FFBS users were commuting, school, and leisure trips in Hangzhou, China.
Moreover, Li et al. (2018) mentioned that FFBS was mainly used for short city trips, especially
for commuting and schooling in Jiangsu, China. Li and Kamargianni (2018) noted that BSSs
are more likely to be selected for leisure trips than commuting trips. Li et al. (2019) stated that
for bike-sharing users, the commute and attending school are the primary trip purpose, followed
by social entertainment and errand, and concluded that non-student users prefer to use PBS for
fixed-purpose trips such as HBW trips. However, students are likely to use FFBS for flexible
travel, such as recreation trips. Noland et al. (2019) mentioned that the trips which start and
end at the same docking stations are primarily recreational.
2.2.6.2.4 Trip distance
Travel distance is an influential factor in bike-sharing usage (Fishman, 2016; Campbell et al.,
2016; Du and Cheng, 2018; Li, 2019). There is a negative correlation between the bike-sharing
ridership rate and the travel distance (between origin and destination) (El-Assi et al., 2017).
Chen et al. (2020) stated that as the travel distance rose, SBBS and FFBS usage reduced. Ji et

60
al. (2020) mentioned that the negative correlation is for both the SBBS and the FFBS; however,
SBBS users are more likely to travel further and longer than FFBS users.
Du et al. (2019) found that the riding distance for FFBS is mostly (80%) between 1 km and
5 km. However, Li et al. (2019) noted that FFBS appeals more to those interested in long-
distance travel. In the study of Du and Cheng (2018) in Nanjing, China, the travel patterns in
FFBS were divided into three categories to detect the influential factors and characteristics of
different travel patterns in FFBS. These three categories were 1) Origin to Destination Pattern
(ODP) (the user uses FFBS to reach the destination directly), 2) Travel Cycle Pattern (TCP)
(origin and destination are the same), and 3) Transfer Pattern (TP) (there is a transfer between
FFBS and other travel modes). Results indicated that residents who travel short distances are
more likely to select TCP and ODP, and when their travel distance reached 4 km, there was a
considerable shift towards TP. In addition, the price affected residents' travel patterns, with
residents showing a tendency to choose FFBS when traveling short distances if they found
FFBS quickly.
2.2.6.3 Bike-Sharing characteristics
One of the most significant factors impacting the demand for bike-sharing is the bike-sharing
characteristics, including travel cost, travel comfort, and helmet provision. The impact of bike-
sharing characteristics is discussed below.
2.2.6.3.1 Travel cost
The cost of BSS tickets is an important factor to consider (Fishman, 2016; Du and Cheng,
2018; Nikitas, 2018). Li et al. (2019) found that changes in the price of BSS at different times
of the day influence its use. For instance, a price reduction could increase the BSS usage from
7:00 am to 10:00 am if the losses from falling prices are less than the gains from raised usage.
A sudden rise in the price of bike-sharing tickets can diminish the level of BSP use for low-
income communities, unlike residents living in middle-income or high-income regions. It
reflects the influence of socio-economic features on BSS (Goodman and Cheshire, 2014).
2.2.6.3.2 Travel comfort
Convenience is the primary factor motivating cycling and contains many facilities, such as
simplicity of payment and membership procedures (Zanotto, 2014; Leister et al., 2018). Also,
the ease of picking up and dropping- off the FFBS can increase demand rates (Li et al., 2019).
In addition to the mentioned general tangible benefits of cycling, bicycle-sharing brings about
a higher level of comfort for users, which can persuade more individuals to adopt cycling for
short trips (Bachand-Marleau et al., 2012). Because of its flexibility, BSS is known as a
convenient means of transportation for short distances and one-way trips (Hyland et al., 2018).
Hence, BSSs are a promising initiative to raise the tendency for cycling among people, that
their advantages to users and society are well known.
2.2.6.3.3 Helmet provision
It is clear that there is an adverse correlation between the use of helmets and BSS demand, and
BSS members' helmet usage rate is less than that of private cyclists (Bonyun et al., 2012;
Kraemer et al., 2012; Grenier, 2013; Fishman et al., 2013; Basch and Zagnit, 2014; Basch et
al., 2014). BSS bikes are usually rented for "unplanned" short-term trips (Fishman, 2016). Also,
mandatory helmet laws reduced bike-sharing demand, and the reason for this may be due to

61
the unwillingness to carry the helmet and not because of wearing it (Fishman et al., 2014). In
addition, Grenier et al. (2013) reported that females (50%) were more likely to wear a helmet
compared to males (44%) in Montreal, Canada. In contrast, Basch and Zagnit (2014) mentioned
that males (52.7 %) used helmets more often than females (41.2%). Besides, Grenier et al.
(2013) noted that youths had more helmet usage levels than young adults, 73% and 34%,
respectively. Also, the helmet-wearing use proportion was higher for commuting trips (58.9
%) versus recreational trips (42.4 %) in New York City, USA.
2.2.6.4 Built environment and land use
Infrastructure and Transportation Facilities, land use, and accessibility factors influence bike-
sharing use.
2.2.6.4.1 Infrastructure and transportation facilities
It is necessary to determine the relationship between BSS usage, built environment, and land
use attributes to comprehend people’s bike-sharing choice behavior (Shen et al., 2018; Duran-
Rodas et al., 2019). Up to the present, many studies have identified the built environment and
land use factor that prevents/promotes the use of BSP (Faghih-Imani et al., 2017; Wang et al.,
2018). It should be noted that the sustainability of the BSS pertains to bicycle network
accessibility and connectivity. Knowing how to allocate resources at the station level is
essential for BSPs. There is also ample evidence that public agencies need to perceive the
temporal effects of bicycle lane investment on bicycle use, especially in smart cities where a
keen understanding of interactions between bike-sharing operators and agencies is imperative
(Chow & Sayarshad, 2014).
As previous research reported, there is a significant positive relationship between the
presence of bike lanes and bike-sharing ridership (Buck and Buehler, 2012; Fishman et al.,
2015). In general, the expansion of bike lane networks near bike-sharing stations is associated
with the desire to cycle more (Krykewycz et al., 2010; Buck and Buehler, 2012; Faghih-Imani
and Eluru, 2016b; Kabak et al., 2018). Besides, the bike-sharing stations placed along the same
high-quality bike routes have higher trip rates than other pairs of stations (Noland et al., 2019).
Also, bicycle lanes raise bike-sharing trips on weekends and holidays and increase casual users'
travel (Noland et al., 2016). Bike-sharing stations, which are close to off-road infrastructure,
are most active in Brisbane, Australia (Mateo-Babiano et al., 2016). Also, extending the length
of off-street bike routes could remarkably promote BSS usage (Wang and Akar, 2019). Mateo-
Babiano et al. (2016) stated that the length of off-road bikeways located within 400m of the
bike-sharing stations strongly correlates to the use of the PBS program. Similarly, Zhou (2015)
employed a flow clustering analysis to specify the optimal distance and reported that the
appropriate value for buffer (service radius of bike share station) distance is 402 m. Besides,
Wang et al. (2018) mentioned that the length of off-road within a 500m station buffer positively
influences the amount of trip generation. Besides, it was mentioned that the length of the
sidewalk does not affect the use of BSP.
Furthermore, bicycle-friendly facilities and concentrated amenities propel many people to
use the BSS when paired with a well-designed public bicycle system (Gleason and Miskimins,
2012). Lu et al. (2018) noted that high-volume unmarked cycling routes reduce BSS usage.
Also, bike-sharing users tend to select routes with separated lanes instead of the shortest routes.

62
Also, according to Jain et al. (2018), casual BSS users are more likely to ride bikes in areas
with separate bike lanes and paths. Xu and Chow (2019) mentioned that installing additional
miles of bike lanes and a more significant number of bike-sharing stations leads to higher bike-
sharing ridership. Wang and Lindsey (2019) found that the length of on-street bike facilities
positively correlates with BSS use. In addition, the impact of bicycle-sharing facility size is
stronger than the influence of bike-sharing access on BSS usage. Wang and Akar (2019) found
that installing bicycle racks positively affects the greater use of BSS. This effect is higher for
females. Specifically, a 1% rise in the number of bike racks is associated with a 1.18%
increment in BSS usage by females. Hence, the transport-related infrastructure plays a
significant role in the bike-sharing users' decision choice (Jennings, 2011; Zanotto, 2014; Ricci,
2015; Faghih-Imani et al., 2017; De Chardon, 2017; Duran-Rodas et al., 2019).
However, De Chardon et al. (2017) mentioned that the system expansions, including
increasing the number of stations, could not improve BSS performance. Also, Wang and
Lindsey (2019) noted that installing new stations in areas without proper bike-share access and
without creating and connecting them as part of a dense network system may not significantly
raise BSS use. According to Shen et al. (2018), a more extensive FFBS fleet leads to greater
use. Nevertheless, as the size of the fleet increase, the marginal effect reduces; hence, the
utilization level of each bike decreases. Also, due to limited public space and road resources,
such growth is not sustainable. Excessive use of the bicycle fleet damages its economic
stability, causes visual pollution, and takes up much public space.
2.2.6.4.2 Land use
Population density and the city's labor market size are prominent indicators contributing to the
bike-sharing trip generation and attraction factors (Hampshire and Marla, 2012; Zhang, 2017).
Duran-Rodas et al. (2019) noted that the city population is important in using SBBSs. Wang
and Lindsey (2019) reported that BSS usage is higher in areas with a higher percentage of retail
land use and a higher population density. Also, Noland et al. (2016) stated that the more
population and employment, the more BSS is used. According to Jain et al. (2018), casual BSS
users are likelier to ride bikes in areas adjacent to tourist hotspots. However, long-term BSS
users often cycle in areas close to high employment density districts. El-Assi et al. (2017) noted
that population density is more decisive for trip generation, while employment density is more
influential for trip attraction. The working point of interest (POI), transit POI, and residential
POI promote using the FFBS and the SBBS (Ji et al., 2020). Noland et al. (2016) found that
areas with higher residential populations were associated with higher subscriber travel rates,
especially on non-working days.
Lin et al. (2020) found that parks can increase bike-sharing usage rates on
weekends/holidays more than on weekdays. In addition, Etienne and Latifa (2014) found that
bike-sharing stations near parks could increase BSS demand on the weekend afternoon in Paris,
France. Also, Duran-Rodas et al. (2019) stated that city leisure facilities are among the factors
influencing the use of SBBSs. Besides, the importance of the influence of some factors is
temporarily different (e.g., the impact of nightclubs during the night). Also, the distance from
a bike-sharing station to car-sharing stations, city centers, memorials, and bakeries affects the
use of the SBBS.

63
Kutela and Kidando (2017) found that the BID in commercial areas is shorter than in
residential land use. A study by Kim et al. (2012) in Goyang, South Korea, indicated that areas
near commercial and residential buildings, parks, schools, and subway stations near the bike-
sharing stations could positively affect bike-sharing. Also, it was observed that on non-rainy
weekdays, commercial buildings could raise public bike usage fifteen times more than
residential buildings; parks attract bike-sharing users three to five times more than subway
stations or schools. Croci and Rossi (2014) identified that the presence of cinemas, universities,
subway and train stations, museums, and limited traffic zones could significantly increase the
levels of SBBS use in Milan, Italy. In contrast, bus and tram stations and theaters have adverse
impacts.
Furthermore, Noland et al. (2016) found that subway stations proximate to bike-sharing
stations lead to a rise in bike-sharing trips. It is worth noting that for both casual and long-term
BSS users, proximity to major transportation hubs is a significant factor (Jain et al., 2018).
According to Lin et al. (2020), the proximity to colleges does not show a noticeable rise in
levels of bike-sharing use. Buck and Buehler (2012) stated that mixed-use planning, in which
two or more residential, institutional, cultural, commercial, and industrial uses are blended, is
essential in encouraging bike-sharing utilization. Hence, planning urban areas with more
diverse economic activities can increase the use of FFBS (Shen et al., 2018).
In a study by Zhao et al. (2019) in Nanjing, China, it was reported that SBBS stations are
prone to unbalanced demand, meaning that SBBSs are facing excessive demand or suffer from
a shortage of parking supply. It was found that the factor of the built environment has a
significant relationship with the number of bike-sharing reallocations. Also, SBBS stations
with the highest number of reallocations are placed close to clinics/hospitals, residences,
employment areas, bus stops, subway stations, amenities, parks, sports facilities, and
restaurants. While stations proximate to educational institutions, hotels, leisure facilities,
entertainment venues, and shopping malls are more likely to have balanced demand and supply.
Besides, the stations' capacity is the most substantial factor in bike reallocation. In addition, it
was revealed that the presence of restaurants and areas with high employment density
positively impact bike removal in the morning and bike refilling in the afternoon at SBBS
stations. Also, Vogel et al. (2011) stated that due to short-term rental and one-way utilization,
imbalances occur in the spatial distribution of bicycles. Therefore, planning the right location
for bike-sharing stations can reduce imbalances. Shen et al. (2018) mentioned that the general
management, optimization, and rebalancing of SBBS are different from FFBS. In order to
rebalance SBBS, it is only required to consider pick-up and drop-off at stations. However, since
FFBS can be parked anywhere where parking is legal, it potentially complicates the rebalancing
of FFBS.
2.2.6.4.3 Accessibility
Accessibility is a considerable factor influencing bike-sharing demand. Bachand-Marleau et al.
(2012) attempted to ascertain the elements which enhance people's tendency to use the shared
bike system and the factors affecting the frequency of use. It was revealed that the location of
shared bicycle stations is an essential factor in using shared bicycles. The home's proximity to

64
docking stations significantly impacts the likelihood of choosing the shared bicycle system.
Hence, the higher the number of docking stations near the origins of potential users in
residential areas, the higher the number of system users. Besides, access to the most proximate
bike-sharing station is needed for pick-up and return activities (Shaheen et al., 2011). Hence,
the bike-sharing stations should be in the nearest locations to gain the maximum coverage and
attract the most significant number of people who desire to rent a bicycle (Dell'Olio, 2011). In
addition, the proximity of bike-sharing stations to each other and the users' position raise the
levels of bike-sharing use (Bachand-Marleau et al., 2012; Rixey, 2013). Also, Wang and
Lindsey (2019) found that relocating old stations or placing new ones can reduce the distance
to the stations, which leads to an improvement in access. Therefore, bike-sharing accessibility
is higher in areas with dense bike-sharing services. Wag et al. (2016) stated that the proximity
of bike-sharing stations to parks, a central business district, waterways such as lakes or rivers,
and access to trails are essential factors in increasing the use of BSS.
In addition, Faghih-Imani et al. (2017) noted that bike station density and average capacity
influenced the rate of BSS use. Faghih-Imani and Eluru (2015) indicated that bike-sharing
members prefer high-density stations with small capacities; however, daily users are likely to
favor fewer capacity stations with more extensive docks. SBBS systems possess many stations
and ready-to-use bikes to quickly pick-up and return bikes in cities (Lin and Yang, 2011). In
order for users to embrace such systems, it is significant to ensure high bicycle availability at
stations (Froehlich et al., 2008; Vogel and Mattfeld, 2011; Feng and Li, 2016; Zhang, 2017).
Also, the empty smart parking unit is needed at the stations to place the rented bicycle.
2.2.6.5 Natural environmental conditions
The hilliness, weather conditions, temperature, seasonal effects, and pollution factors are
natural environmental conditions that can influence the demand for bike-sharing.
2.2.6.5.1 Hilliness
The steep slopes can make the ascents difficult for cyclists as the required power for cycling
rises in proportion to the hill's gradient. Also, the descents can trigger unsafe high-speed and
reduce levels of perceived safety for users (Frade and Ribeiro, 2014). In general, the use of
BSS can be reduced when cycling uphill (Jennings, 2011; Bordagaray et al., 2016), especially
when the slope is above 4% (Lu et al., 2018). Fricker and Gast (2016) suggested that reward
policies could incentivize users to return bikes to stations to boost the bike-sharing usage rate
and address the bike-sharing rebalancing problem at uphill stations.
2.2.6.5.2 Weather conditions
Moreover, one of the issues that affect user ridership choice behavior is the impact of weather
conditions on cycling (Simons et al., 2013; Gebhart and Noland, 2014; Shen et al., 2018).
Caulfield et al. (2017) noted that in favorable weather conditions, the travel time and the
number of trips are higher in Cork, which is a small city in Ireland. On the other hand,
theoretically, rainfall, colder weather, high wind speed, and extreme heat are negatively
correlated with levels of bike-sharing use (Corcoran et al., 2014; Gebhart and Noland, 2014;
Faghih-Imani and Eluru, 2016a; Hyland et al., 2018; Kim, 2018; Sun et al., 2018). Daily usage
is reduced due to wind speed, snowfall, and rain (Lin et al., 2020). De Chardon et al. (2017)
showed that wind could negatively influence BSS performance. According to Martinez’s

65
(2017) study, lower wind speeds, precipitation rates, higher temperatures, and less snowfall
can raise BSS usage in New York City. A study by Reiss and Bogenberger (2016) in Munich,
Germany, stated that rainfall could reduce the use of bicycle-sharing by much less than the
average, and the demand for bike-sharing trips returns to its average level 3 hours after heavy
rains. Eren and Uz (2020) noted that rain is the most unfavorable weather condition that impacts
the use of bicycles.
Gebhart and Noland (2014) utilized a rich dataset encompassing hourly usage information
and weather patterns to survey the effect of weather on the frequency and the levels of bike-
sharing use in Washington, DC, United States of America. Results showed that rainfall has
more impact on the rate of bike-sharing trips when the bike-sharing stations are proximate to
the subway stations compared to when the bike-sharing stations are far from the subway
stations. Moreover, on rainy days, the number of trips is about 0.56 times less, and the travel
time is around 2.8 min shorter than on non-rainy days. In that study, the average wind speed
was 13.2 km/h, defined as a "gentle breeze." It was determined that increasing wind speeds
reduces the number of bike-sharing trips and trip duration as people are less willing to cycle
on windy days. Besides, it was specified that the impacts of fog, snowfall, and thunderstorms
are not statistically significant for either the number of trips made or the duration of the trip.
Also, it was found that fog and thunderstorms could raise the trip duration for registered users
in the BSS (0.2% and 4.4%, respectively). Still, it significantly reduced the trip duration for
casual users (36.1% and 29.3%, respectively). For registered users, trip durations declined by
9.4% in snow and 10.1% in the rain. Reduced trip duration for casual users in these weather
conditions were much higher, 12.1% in snow and 22.4% in the rain. It was identified that if the
subway station is available as an alternative transport mode, the rainy days, the temperature
between -6.7° C and -1.7° C, and the absence of adequate daylight can cause a significant
reduction in the number of bike-sharing trips.
2.2.6.5.3 Temperature
The impact of temperature on bike-sharing use has been studied as one of the most important
factors. There is a positive relationship between bike-sharing demand and temperature rise
(Faghih-Imani et al., 2014; Hyland et al., 2018). Eren and Uz (2020) mentioned that bike-
sharing trip production positively correlates with the temperature when the temperature is
between 0–20 °C. Also, temperatures of 20 to 30 °C without precipitation raise the likelihood
of using the BSS. Wang et al. (2018) examined the impact of weather conditions on demand
for BSSs in different age cohorts in New York City. The demand for bike-sharing for all age
groups was positively related to the temperature of 12 to 16 °C. However, this demand
negatively correlated with the weather temperature of 27–32 °C. Also, temperatures of 21–27
°C adversely impacted the demand for bike-sharing among young people between the ages of
16 and 27 but had a positive effect among other age groups. In Kim's (2018) research in
Daejeon, South Korea, temperatures above 30 °C were considered "scorching heat." Because
on only 49 days of the year, the temperature in 2015 was above 30 °C. It was found that when
the temperature is above 30 °C, the use of the BSS is reduced. In contrast, El-Assi et al. (2017)
noted that the demand for bike-sharing in Toronto, Canada, where the temperature can reach
42 °C, is rising at temperatures above 30 °C. Similarly, Jing and Zhao (2015) showed that the

66
best temperature at which the demand for bicycle sharing reaches its maximum is between 30
°C and 35 °C in Washington, DC, USA. However, Lin et al. (2020) found that temperature was
not linearly related to bike-sharing daily use in Beijing, China.
Proper ambient temperature is associated with positive changes in physical activity
participation (Faghih-Imani and Eluru, 2016a). As Cycling is an important and widely used
form of physical activity, Heaney et al. (2019) opted to elucidate the relationship between
ambient temperature and levels of bike-sharing use. Also, it was investigated how rising
ambient temperatures caused by climate change might affect active transportation in New York
City. The results showed that the highest total hours of bike-sharing use and the maximum
average distance traveled in the year's warm months (March-October) occurred. Although the
levels of bike-sharing use are positively related to higher temperatures, bike-sharing use is
reduced when the temperature is above 26°C–28°C in New York City. Also, Because of climate
change, bike-sharing use might increase by up to 3.1% by 2070. In addition, in the future, the
use of BSSs will increase in winter, spring, and fall. This projected increase outweighs the
reductions which occur in the rate of bike-sharing use in summer. It should be stressed that
although the use of the BSS in New York City is expected to increase by the middle of this
century, this trend may be reversed if the temperature continuously rises.
The results of a study by Gebhart and Noland (2014) revealed that colder weather, rain,
and high humidity decrease the trip duration and the probability of using the BSS. Further,
most trips seem to be made when the temperature is between 26.7° C and 31.7° C. Besides,
When the temperature is between -12.2° C and 4.4° C, the average trip duration is shorter,
unlike when the temperature is in the range of 10° C to 15° C. Also, temperature between 21.1°
C and 31.7° C was significantly associated with increased travel length. However, there was
not any considerable difference between the trip duration of the temperature above 32.2 and
the trip duration between 10° C and 15° C. Plus, it was found that a change of 1% in average
humidity (63.63%) can reduce the travel frequency by 0.94%. Besides, it is usually assumed
that 32.2° C to 37.2° C is not favorable for cycling. Surprisingly, it was found that the number
of trips at temperatures of 32.2° C to 37.2° C has increased dramatically. Therefore, increasing
humidity reduces bike-sharing trips, but high temperatures are not necessarily the case.
2.2.6.5.4 Seasonal effects
Moreover, there are seasonal effects, meaning most bike-sharing trips are made in the summer,
while bike-sharing demand is relatively low in the spring and fall. Godavarthy and Taleqani
(2017) observed that BSS usage in winter was equal to 10%-30% of its peak use in summer in
the cold cities of the United States. Similarly, Rudloff and Lackner (2014) stated that demand
for bike-sharing declines significantly in the winter, even at the heavily used stations in Vienna,
Austria. In addition, Kutela and Kidando (2017) pointed out that the likelihood of long BID
rises in the winter, especially when it snows and rains. Hyland et al. (2018) reported that
members who use the BSS in the morning are less affected by winter weather than other users.
These members are less likely than tourists or casual users to be influenced by snow and cold
weather and rely on BSSs to improve their commute. Also, in terms of sensitivity to
unfavorable weather, Fournier et al. (2017) noted that recreational cyclists are more sensitive
to unfavorable weather than their daily commuting counterparts.

67
Miranda-Moreno and Nosal (2011) examined the sensitivity of cyclists to weather
conditions in Montreal, Canada. It was observed that a sharp temperature rise could reduce the
use of the BSS in warm months and raise bike-sharing usage in cold months. Cyclists seem
more sensitive to low temperatures, regardless of whether the average temperature is cold or
hot. Although a sharp drop in temperature can reduce bike-sharing usage in colder months, it
can raise the usage of the bike-sharing level in warmer months. Furthermore, it was found that
a rise of 10% in temperature from 14.7 °C causes an average increase of 4%-5% in the hourly
volume of the BSS. Additionally, it was mentioned that the bike-sharing volume raised from
32% to 39% in the summer. However, when the humidity exceeds 60%, and the temperature
exceeds 28 °C, the bike-sharing demand decreases.
2.2.6.5.5 Pollution
In a study by Lin et al. (2020), it was stated that heavy pollution and light do not significantly
affect bike-sharing use; however, severe pollution adversely impacts bike-sharing usage in
Beijing, China. Moreover, Li and Kamargianni (2018) realized that air pollution significantly
negatively affected bike-sharing choices in Taiyuan, China. Nonetheless, improving BSS
services, such as saving on access time and travel costs, is more effective in raising BSS use
than improving air quality.
2.2.7 Summary
Factors such as natural environmental conditions, built environment and land use, trip-related
characteristics, bike-sharing characteristics, and socio-demographic characteristics influence
bike-sharing use. Table 24 indicates these factors and their sub-factors effects on the use of
bike-sharing. Table 24: Factors affecting bike-sharing choice.
Factors
Sub-factor
Positive impact
Negative
impact
Reference
(studied the SBBS)
References
(Studied The
FFBS)
Natural
environmental
conditions
Weather
condition
Favorable weather
Adverse
weather such
as rainfall, high
humidity
Miranda-Moreno and Nosal,
2011; Gebhart and
Noland, 2014; Corcoran et al.,
2014; Faghih-Imani and Eluru,
2016a; Caulfield et al., 2017;
Fournier et al., (2017); De
Chardon et al., 2017; Martinez,
2017; Kutela and Kidando,
2017; Sun et al., 2018; Lin et
al., 2020; Eren and Uz, 2020
Reiss and
Bogenberger,
2016: Shen et al.,
2018
Temperature
Warm temperature
Cold
temperature
Miranda-Moreno and Nosal,
2011; Faghih-Imani et al.,
2014; Corcoran et al., 2014;
Gebhart and
Noland, 2014; Jing and Zhao,
2015; Faghih-Imani and Eluru,
2016a; El-Assi et al., 2017;
Martinez, 2017; Wang et al.,
2018; Hyland et al., 2018;
Kim, 2018; Heaney et al.,
2019; Eren and Uz, 2020; Lin
et al., 2020
Seasonal effect
Summer
Winter
Rudloff and Lackner, 2014;
Godavarthy and Taleqani,
2017; Kutela and Kidando,
2017; Heaney et al., 2019;
Eren and Uz, 2020

68
Factors
Sub-factor
Positive impact
Negative
impact
Reference
(studied the SBBS)
References
(Studied The
FFBS)
Pollution
Severe
pollution
Li and Kamargianni, 2018
Hilliness
Steep roads
Jennings, 2011; Frade and
Ribeiro, 2014; Fricker and
Gast, 2016; Bordagaray et al.,
2016
Lu et al., 2018
Built
environment and
land use
Land use
Living near a
densely populated
community, route,
retail density,
commercial
buildings, leisure
facilities, and
presence of parks
along the journey
Buck and Buehler, 2012; Kim
et al., 2012; Hampshire and
Marla, 2012; Bachand-
Marleau et al., 2012; Noland et
al., 2016; Croci and Rossi,
2014; Etienne and Latifa,
2014; Noland et al., 2016;
Zhang, 2017; El-Assi et al.,
2017; Kutela and Kidando,
2017; Jain et al., 2018; Duran-
Rodas et al., 2019; Wang and
Lindsey, 2019; Zhao et al.,
2019; Duran-Rodas et al.,
2019; Lin et al., 2020; Ji et al.,
2020
Shen et al., 2018;
Ji et al., 2020
Accessibility
Living within a
proximate distance
to public transit
stations and
proximity of
docking stations to
residential
Housing
Bachand-Marleau et al., 2012;
Kim et al., 2012; Wang and
Lindsey (2019); Zhao et al.,
2019; Ji et al., 2020
Bike-sharing
station distance
from major roads
Station proximity
to major roads
Zhou, 2015; Mateo-Babiano et
al., 2016; Wang et al., 2018;
Wang and Akar, 2019; Noland
et al., 2019
Bike-sharing
station distance
from transit
stops
Station proximity
to transit stops
Croci and Rossi (2014);
Noland et al., 2016; Jain et al.,
2018; Duran-Rodas et al.,
2019; Zhao et al., 2019
Bike-sharing
station distance
from bicycle
lanes
Station proximity
to bicycle lanes
Krykewycz et al., 2010; Buck
and
Buehler, 2012; Fishman et al.,
2015; Noland et al., 2016;
Faghih-Imani and Eluru,
2016b; Jain et al., 2018; Kabak
et al., 2018; Xu and Chow,
2019
Lu et al., 2018
Fleet size
The higher number
of stations, larger
size, and length of
the facility
Bachand-Marleau et al., 2012;
Faghih-Imani and Eluru,
2016b; Wang and Lindsey,
2019; Xu and Chow, 2019;
Wang and Akar, 2019
Shen et al., 2018
Bike-sharing
design
Satisfy with the
design of shared
bikes
Bachand-Marleau et al. (2012)
Bike-sharing
station distance
from other bike-
sharing stations
Station proximity
to other bike-share
stations
Bachand-Marleau et al., 2012;
Rixey, 2013
Trip-related
characteristics
Trip purpose
Commuting,
traveling to school,
leisure trips
Fishman, 2016; Li and
Kamargianni, 2018; Li, 2019;
Noland et al., 2019; Li et al.,
2019; Chen et al., 2020
Li et al. (2018); Li
et al. 2019; Chen et
al., 2020
Trip distance
High travel
distance
Fishman, 2016; Campbell et
al., 2016; El-Assi et al., 2017;
Li, 2019; Li et al., 2019; Ji et
al., 2020; Chen et al. (2020)
Li et al., 2018; Du
and Cheng, 2018;
Du et al. 2019; Li
et al., 2019; Ji et
al., 2020; Chen et
al., 2020

69
Factors
Sub-factor
Positive impact
Negative
impact
Reference
(studied the SBBS)
References
(Studied The
FFBS)
Departure time
Peak hours
Froehlich et al., 2008;
Kaltenbrunner et al., 2010;
Kim et al., 2012; Faghih-Imani
et al., 2014; O’Brien et al.,
2014; Corcoran et al., 2014;
Reiss and Bogenberger, 2016;
Ahillen et al., 2016; Mateo-
Babiano et al., 2016; Kutela
and Kidando, 2017; Kim et al.,
2018; Heaney et al., 2019; Li et
al., 2019; Ji et al., 2020; Lin et
al., 2020
Zhang and Mi,
2018; Li et al.,
2019; Ji et al.,
2020
Travel time
Short trips
Jensen et al., 2010; Buehler
and Hamre, 2014; Mateo-
Babiano et al., 2016
Bike-sharing
characteristics
Travel cost
Low travel cost
High travel
cost
Goodman and Cheshire, 2014;
Fishman, 2016; Nikitas, 2018;
Li and Kamargianni, 2018
Du and Cheng,
2018; Li et al.,
2019
Helmet
provision
Mandatory
helmet laws
Bonyun et al., 2012; Kraemer
et al., 2012; Grenier, 2013;
Fishman et al., 2013; Fishman
et al., 2014; Basch and Zagnit,
2014; Basch et al., 2014;
Fishman, 2016
Travel comfort
Comfort
Zanotto, 2014; Leister et al.,
2018
Socio-
demographic
characteristics
Age
Younger users
Fuller et al., 2011; Vogel et al.,
2014; Zanotto, 2014; Fishman
et al., 2015; Ricci, 2015; Raux
et al., 2017; Nikitas, 2018;
Wing et al., 2018; Li et al.,
2019; Eren and Uz, 2020;
Chen et al., 2020
Du and Cheng,
2018; Li et al.,
2019; Chen et al.,
2020
Gender
Male users
Ogilvie and Goodman, 2012;
Zanotto, 2014; Vogel et al.,
2014; Goodman and Cheshire,
2014; Buehler and Hamre,
2014; Ricci, 2015; Faghih-
Imani and Eluru, 2015;
Fishman et al., 2015; Fishman,
2016; Raux et al., 2017;
Nikitas, 2018; Wang and Akar,
2019; Li et al., 2019; Chen et
al., 2020
Du and Cheng,
2018; Li et al.,
2019; Chen et al.,
2020
Occupation and
economic status
Higher-income,
affluent people
Low income
Maurer, 2011; Rixey, 2013;
Zanotto, 2014; Fishman et al.,
2015; Ricci, 2015; Murphy and
Usher, 2015; Raux et al., 2017;
Li et al., 2019
Li et al., 2018; Li
et al., 2019
Residence status
(permanent
residence or not)
Permanent
residence
Li et al., 2019; Du et al., 2019
Li et al., 2019; Du
et al., 2019
Ownership
status
(bike, car,
scooter)
Bachand-Marleau et al., 2012;
Shaheen and Guzman, 2011;
Fishman et al., 2015; Chen et
al., 2020
Du et al., 2019;
Chen et al., 2020
Education level
Well-educated
background
Fuller et al., 2011; Bachand-
Marleau et al., 2012; Zanotto,
2014; Fishman et al., 2015;
Ricci, 2015; Li et al., 2019;
Cheng et al., 2020
Du and Cheng,
2018; Li et al.,
2019; Cheng et al.,
2020
It is worth mentioning that although most of the factors affecting bicycle use and bike-
sharing have similar effects, there are differences in some of them. For instance, the purpose

70
of bicycle travel is primarily recreational trips. However, the purpose of bike-sharing trips is
in the broader area, including commuting, school trips, and leisure trips. Moreover, people with
higher incomes and affluent people use bike-sharing more than people with low incomes. On
the other hand, low-income people ride bicycles more than high-income people. Besides,
owning a car does not seem to reduce the likelihood of using bike-sharing. However, it can
reduce the use of bicycles.
2.3 An overview of scooter-sharing
The scooter-sharing system and its benefit are explained in Chapter 1. This section provides an
overview of scooter-sharing services to figure out better the important criteria and sub-criteria
that can impact the use of scooter-sharing. In this regard, a brief history of e-scooter-sharing,
advantages and disadvantages of e-scooters, e-scooter vs. other transport modes, and factors
affecting demand for e-scooters and its summary are noted as follows.
2.3.1 A brief history of e-scooter-sharing
In 2017, Bird and Lime (American transportation companies) introduced dockless electric kick
scooters, which are a modern means of transportation (micro-mobility) (Almannaa et al., 2020).
In Europe's case, the most significant interest in scooter-sharing services occurred in 2018,
when these systems began operating in Europe's largest capitals (Turoń and Czech, 2019). This
trend soon spread to several cities in the USA and around the world to various European
countries, Canada, Central and South America, Australia, New Zealand, and so on (Sipe and
Pojani, 2018; Choron and Sakran, 2019; Petersen, 2019; Shaheen et al., 2020). E-scooters are
considered the newest means of transport in the evolving sharing economy (Popov and Ravi,
2020). This new mobility solution is becoming more popular with shared mobility operators
and new social trends (Turoń and Czech, 2019). It seems that e-scooters can meet instant
demand (Gössling, 2020). With the rapid growth of the on-demand and sharing economy, the
scooter-sharing market has accelerated rapidly over the past year, and cities worldwide host
scooter-sharing activities (He and Shin, 2020). The increasing use and acceptance of shared e-
scooter services reflect the untapped demand for innovation in urban mobility, representing
another disruptive force in transport services. Besides, most e-scooter-sharing users were not-
regular, while 22% utilized the e-scooter-sharing service several times a month in Europe and
North America (Popov and Ravi, 2020).
2.3.2 General advantages and disadvantages of e-scooters
E-scooters have some pros and cons. The main ones are as follows.
2.3.2.1 Advantages
• Provide an additional transportation solution that allows users to address the first/last
mile issue (Turoń and Czech, 2019),
• A sustainable alternative to fossil fuels cars (Carrese et al., 2021),
• Weaving through dense traffic (Sanders et al., 2020),
• Contribute to reducing traffic congestion (Carrese et al., 2021),

71
• Additional transportation solutions enhance the attractiveness of tourism in the urban
environment where e-scooter-sharing is located (Turoń and Czech, 2019),
• Positive effect on the environment because of the e-mobility and negative noise
reduction (Turoń and Czech, 2019),
• Low maintenance costs (Turoń and Czech, 2019),
• Relatively low cost of purchasing a scooter (Turoń and Czech, 2019),
• Education for e-mobility because of the high availability of e-scooters across the entire
society (Turoń and Czech, 2019),
• It requires less physical effort than cycling or walking (Younes et al., 2020).
According to Popov and Ravi (2020), providers of e-scooter-sharing services promote e-
scooters as a better option than cars for environmental reasons. Hence, service providers
promote e-scooter usage by influencing the customers to believe that they are making the right
decision by utilizing an e-scooter-sharing service and contributing to the carbon-free
transportation mode. Also, promoting e-scooter-sharing as an environmentally friendly option
can raise service loyalty. Most e-scooter-sharing service users (millennials) consider using their
e-scooters to be environmentally friendly and recognize that sharing services are sustainable
and reflect a modern lifestyle.
2.3.2.2 Disadvantages
• Low speed compared to car or bus (Turoń and Czech, 2019),
• Need to charge (Turoń and Czech, 2019),
• It cannot carry more than one person (Turoń and Czech, 2019),
• Limited load carrying capacity, such as difficulty in carrying luggage (Turoń and
Czech, 2019),
• A short lifespan (Moreau et al., 2020),
• The problem of parking them on the sidewalk (James et al., 2019),
• Accidents and injuries (Schlaff et al., 2019).
E-scooter-sharing is expanding significantly and can reduce traffic congestion in dense
cities. Nevertheless, this new micro-mobility transport mode creates many operational, privacy,
and safety concerns (Li et al., 2020). Immediately after the deployment of e-scooters, there
were complaints from non-users, especially pedestrians, who felt another violation in their
public space (Tuncer and Brown, 2020). Some users park their e-scooter without following the
traffic rules. They leave the e-scooter in positions and places that dramatically reduce urban
space and interfere with pedestrians and other vehicles. In order to counter poor parking and
increase the popularity of e-scooters among city dwellers, some agents have been hired by e-
scooter-sharing companies. Their main task is to reposition e-scooters at short distances to
eliminate inappropriate and irregular parking created by users and ensure urban decoration
(Carrese et al., 2021).
James et al. (2019) surveyed 181 users and non-users of e-scooters and examined their
perceived safety for e-scooters users and the experiences of scooter-blocked sidewalks in
Virginia, USA. It was found that there were highly divergent responses about safety and

72
perception of sidewalk blockage. It was also demonstrated that 16% of respondents noticed
that the e-scooters were not appropriately parked, and 6% of the e-scooters blocked the
pedestrian crossing. In contrast, Fang et al. (2018) reported that most scooters were well parked
in downtown San Jose, USA. Additionally, less than 2% of scooters blocked access for the
disabled. Of the scooters parked on the sidewalks, 90% are not parked in pedestrian traffic.
Most did not obstruct pedestrian traffic, even among the 10% of scooters parked on the
sidewalk that was not in the street furniture zone or sidewalk edge. Importantly, more secure
infrastructure and lower street speed limits reduce sidewalks' illegal use (Shaheen and Cohen,
2019). The almost spontaneous proliferation of e-scooters has prompted e-scooter-sharing
companies and the government to address issues partly due to concerns about the large number
of e-scooters entering vehicle traffic. These issues are affected by the e-scooter users' decisions
and behaviors that, despite being licensed to drive passenger vehicles, have potentially limited
experience with an e-scooter in traffic (Todd et al., 2019). Hence, the complexities of micro-
level interactions in macro-level decision-making have to be considered by governments
(Gibson, 2020). Municipal governments have enacted e-scooter regulations to raise riders’ and
pedestrians' safety, prevent visual pollution, and ensure safety, management, and operation
(Anderson-Hall et al., 2019; Almannaa et al., 2020). Safety, promoting equitable access to
services, assessing the effects of e-scooters on traffic, and sustainability are the primary
purposes that most cities focus on (Clewlow, 2019). Urban planners should cautiously
introduce maximum speed, mandatory use of bicycle infrastructure, and private parking and
limit authorized operators' numbers (Gössling, 2020). It is worth noting that differences in city
size, climate, geography, and other characteristics may lead to different policies and approaches
(Riggs and Kawashima, 2020).
E-scooters face safety challenges due to increased vibrations, speed changes, and limited
ride environments (Ma et al., 2021). Therefore, safety is of paramount significance on a shared
footpath. While lower riding speed can decrease the likelihood and severity of injuries, the
speed of e-scooters may not be the only factor in assessing perceived risk on a shared footpath.
Because of feeling safe, pedestrians had a similar perception of the speed of 10 km/h or 15
km/h (Che et al., 2020). For instance, in Oslo, Norway, one in ten e-scooter users have had an
accident. Most users (46%) feel safe in traffic. However, one in four pedestrians and cyclists
feels unsafe interacting with e-scooter users (Berge, 2019). In Portland, USA, 83% of e-
scooter-related injuries did not involve other means of transportation, 13.6% involved a motor
vehicle, and 2.8% related to a pedestrian (Shaheen and Cohen, 2019). Also, only one collision
(0.6%) involved two scooters. In Brisbane, Australia, not wearing a helmet, consuming alcohol,
and speeding more than 30 km/h were essential factors in e-scooter accidents (Haworth and
Schramm, 2019).
The introduction of the e-scooter-sharing service has created a new injury risk. In Australia,
dislocations or fractures were observed in 32% of patients and 26% with head injuries, one of
which was severe. In addition, isolated partial musculoskeletal injuries were seen in 46% of
patients (Beck et al., 2020). Since the launch of e-scooter-sharing in Salt Lake, USA, a
substantial increase in e-scooter-related trauma has been seen. Of note is the number of patients
with major musculoskeletal and head injuries (Badeau et al., 2019). Similarly, in Brisbane,

73
Australia, abrasions/contusions and dislocations/fractures were the most common injuries
(Mitchell et al., 2019). In Los Angeles, USA, 11% of the injured patients were under 18 years
old, and only 4% of the users have documented the use of helmets (Trivedi et al., 2019). Also,
head and face injuries in Dallas, USA, accounted for 58% of all injuries. The prevalence of
extremity injuries indicates that patients fell off the e-scooter when they had an accident. In
addition, wearing a helmet can decrease craniofacial trauma associated with e-scooters (Trivedi
et al., 2019). In the USA, approximately 87% of emergency visits were for patients undergoing
treatment and discharge. Besides, roughly 15% of injuries related to e-scooters occurred on the
face, ankles, head, knees, and low leg.
Moreover, about 45% of injuries occurred in people aged 10-29. Further, of the 51 million
person-trips taken by e-scooters, 346 injuries per million trips were reported. However, of 4.7
billion person trips taken by bikes, 114 injuries/million trips were reported. The most dangerous
behavior for e-bike and bike cyclists was cycling against the traffic flow in a naturalistic
environment. For e-scooter users, it was riding without a helmet (Watson et al., 2020). A study
at Auckland City Hospital in New Zealand also identified an increased need for urgent
radiology imaging in the first two months after the e-scooters launch (Mayhew and Bergin,
2019).
The emergence of many e-scooters in urban traffic leads to many legal and safety issues.
There are problems with moving and parking e-scooters on the streets, pavements, and
intersections (Turoń and Czech, 2019). In the usage phase, the user's behavior affects
operational safety, particularly compliance with the applicable rules. Besides, dropping off the
scooter is of particular significance for all traffic users' safety in the vehicle's final stage. The
mobile app supporting users in vehicle performance affects system safety (Tubis et al., 2019).
In Brisbane, Australia, roughly 44.7% of shared e-scooter users rode illegally, such as double-
riding (2%), not wearing a helmet (35.8%), and riding on the road (6.9%). The correct use of
the helmet in e-scooter-sharing was lower than in bike-sharing, 60.9% and 81%, respectively
(Haworth and Schramm, 2019). Therefore, policies must be adopted to reduce e-scooter-related
injuries, including lower speed limits, night-time curfew, zero blood alcohol concentrations,
and helmet use (Brownson et al., 2019; Bloom et al., 2021). Also, e-scooters require curb space
management because they share public right-of-way with other transport modes, such as
pedestrians on the sidewalk (Ma et al., 2021).
2.3.3 E-scooter vs. other transport modes
E-scooter-sharing has been hailed as an alternative to personal motor transportation, primarily
cars, by urban transportation planners (Gössling, 2020; Caspi et al., 2020). Some e-scooter-
sharing users in Portland, USA, replaced the motor vehicle with e-scooter-sharing. E-scooter-
sharing has also replaced low-emission active transport trips (Shaheen and Cohen, 2019). Also,
e-scooters replaced walking and public transportation in Oslo, Norway (Berge, 2019). Besides,
e-scooters can replace up to 1% of taxi trips in Manhattan, USA (Lee et al., 2019). It is
important to notice that e-scooter-sharing can be paired with other mobility modes, especially
public transport (Schellong et al., 2019). For example, e-scooters can increase access to

74
employment centers in Chicago, USA. Compared to the number of job opportunities currently
only available through walking and public transportation, e-scooters can make approximately
16% more jobs (reachable within 30 min) accessible. Besides, for short trips between 0.5 and
2 miles, e-scooters can be a new alternative to the private car (Smith and Schwieterman, 2018).
A Toronto study found that 21% of people would like to consider e-scooters for some of
their current travels, and most would replace their walking (60%) and transit (55%) travels with
e-scooter-sharing (Mitra and Hess, 2021). In the USA, e-scooter-sharing expands
transportation options, creates a car-free lifestyle, and is a viable alternative to private cars or
ride-hailing services for short travels (Clewlow, 2019). In addition, e-scooters can complement
public transportation. By providing a joint service of local public transportation and e-scooter-
sharing, e-scooter-sharing can be promoted as a complementary option rather than an
alternative to public transportation (Severengiz et al., 2020). Furthermore, with motorized and
dockless features, dockless e-scooter-sharing provides more comfortable and faster first/last
mile connections in the city than conventional bicycle-sharing (He and Shin, 2020).
2.3.4 Factors affecting demand for e-scooters
The elements impacting the usage rate need to be identified to better view the demand for e-
scooter. In the literature, the natural environmental conditions, built environment and land use,
trip-related characteristics, scooter-sharing characteristics, and socio-demographic
characteristics have been considered important factors affecting e-scooter-sharing demand.
2.3.4.1 Socio-demographic characteristics
The socio-demographic features, including ownership status, occupation and economic status,
age, education level, and gender of users, affect e-scooter usage.
2.3.4.1.1 Gender
Gender factor plays an essential role in the e-scooter usage rate. In Vienna and New Zealand,
e-scooter-sharing users are mostly male (Laa and Leth, 2020; Curl and Fitt, 2020). In Brisbane,
Australia, males accounted for 75.6% of e-scooter-sharing users (Haworth and Schramm,
2019). Similarly, in Austin, USA, males were more likely than females to travel on e-scooters
(Jiao and Bai, 2020). Also, in Oslo, Norway, the percentage of using e-scooter by males is
higher than females, 44% and 28%, respectively (Berge, 2019). It is important to state that
females might feel more secure when using e-scooters. This may be because they are smaller
than males and can easily ride e-scooters on sidewalks. Besides, females are less likely than
males to cycle long distances. E-scooters enable them to travel long distances more
comfortably. Because females are more likely to wear clothes like skirts, making it easier to
stand on the e-scooter than on bikes (Clewlow, 2019).
2.3.4.1.2 Age
The age of e-scooter users affects the level of e-scooters usage. In Vienna and New Zealand,
e-scooter-sharing users are primarily young (Laa and Leth, 2020; Curl and Fitt, 2020).
Similarly, in Brisbane, Australia, most e-scooter-sharing program users (89.2%) were adults
(Haworth and Schramm, 2019). Also, 10.8% of shared e-scooter users were under 18 years old.
However, this figure was 2% for shared bike users. In addition, most e-scooter users in Oslo,

75
Norway, are under 30 (Berge, 2019). Likewise, the relationship between e-scooters usage and
the percentage of young people in Minneapolis, USA, was significantly positive (Bai and Jiao,
2020). Most e-scooter-sharing users belong to the millennial generation, precisely 20 to 30
years (Popov and Ravi, 2020). Surprisingly, a study conducted in Austin, USA, found that the
proportion of residents under 25 in a neighborhood and the use of e-scooters were negatively
correlated (Jiao and Bai, 2020). In Portland, USA, younger adults positively perceived e-
scooter-sharing. It should be pointed out that younger adults (under 35) are most concerned
about illegally parked and dangerous scooters. However, the elderly (55 years and older) were
most concerned about riding on the sidewalk (Shaheen and Cohen, 2019). Hence, generally,
young people are the most frequent users of e-scooter-sharing. The reason is their lifestyle and
priorities (Rahimuddin et al., 2020). Also, the elders cannot simply use e-scooters (Clewlow,
2019).
2.3.4.1.3 Education level
Education level is also a significant factor affecting e-scooter usage. Highly educated people
are encouraged to use e-scooters in Austin, USA (Jiao and Bai, 2020). Also, in Vienna, most
e-scooter subscribers are highly educated (Laa and Leth, 2020).
2.3.4.1.4 Occupation and economic status
Household income level is another factor that can affect the use of e-scooter-sharing. Generally,
low-income households have a positive impression of e-scooters (Shaheen and Cohen, 2019).
For example, low-income households were more likely to generate e-scooter travel in Austin,
USA (Jiao and Bai, 2020). Also, e-scooters usage correlates with areas with high employment
rates in Austin, Texas. Also, e-scooters are used by students who are likely to have lower
incomes but are not socio-economically low. The lower the income rate in the area, the more
departures and arrivals are made in the morning on weekdays (Caspi et al., 2020). Overall, e-
scooters may have higher acceptance rates by low-income groups and can potentially help cities
achieve justice purposes (Clewlow, 2019).
2.3.4.1.5 Vehicle Ownership
In Europe and North America, about 79% of the e-scooter-sharing scheme users did not possess
an e-scooter; 12% owned an e-scooter. Almost 9% of the users did not have an e-scooter, but
they considered purchasing an e-scooter in the future (Popov and Ravi, 2020). In Portland,
USA, 6% of the local users had sold a vehicle, and only 16% of users have considered selling
a vehicle as they used e-scooter-sharing in Portland, USA. Unlike purchasing a car, acquiring
an e-scooter is relatively inexpensive and easy for people (Popov and Ravi, 2020). In addition,
the rental price of e-scooters is high for sharing services. As a result, some people are eager to
purchase the e-scooter. Hence, there are many separate scooters on the streets and e-scooter-
sharing. This type of personal transportation is called "personal transportation" (Turoń and
Czech, 2019). Hence, the benefits of owning an e-scooter may undermine loyalty to the e-
scooter-sharing service. However, owning an e-scooter does not significantly impact service
loyalty. On the other hand, not owning an e-scooter has advantages, such as the social benefits
of sharing and sustainability (Popov and Ravi, 2020).

76
2.3.4.2 Trip-Related Characteristics
Trip-related characteristics contain Coronavirus Disease (COVID-19), travel time, departure
time, travel distance, and trip purpose impact using e-scooter-sharing.
2.3.4.2.1 Travel time
E-scooter-sharing users seem somewhat more sensitive to travel time than station-based bike-
sharing subscribers (Younes et al., 2020). With high accessibility, usability, and little waiting
time, using an e-scooter on daily trips can save time (Berge, 2019). Traveling short distances
of 3 miles (4.83 km) or less using an e-scooter is faster than driving a car or utilizing a ride-
hailing service in many urban areas of the United States (Clewlow, 2019). In California, USA,
the average travel time with an e-scooter may be less than that with a shared e-bike. The
maximum legal speed of an e-scooter is 15 miles per hour (24.14 kph). This speed is similar to
a cyclist’s traveling on flat terrain and almost twice the average speed of individuals who ride
a bike (regular and electric) through the bike-sharing program (Todd et al., 2019).
2.3.4.2.2 Travel distance
For many people, e-scooters are a fun and convenient way to travel short distances (Gössling,
2020). The average distance traveled per trip is approximately 1.5 miles (about 2.41 Km) (Todd
et al., 2019). Approximately 35% of all personal trips cover distances of less than 2 km, and
75% are less than 10 km. Also, e-scooters were usually used for trips of 0.5 km to 4 km,
equivalent to 5 to 45 minutes of walking (Schellong et al., 2019). In Berlin, Germany, e-
scooters are mainly used for short distances, with an average distance of 1.54 km (Wüster et
al., 2020). It is worth noting that passengers traveling between half a mile (about 800 m) and
two miles (around 3.22 km) receive the most out of e-scooters. Longer scooter trips, especially
trips of more than three miles (approximately 4.83 km), are usually too expensive and
impossible to afford for ordinary city travelers. Most travelers who travel more than three miles
use scooters to access bus and train stations (Smith and Schwieterman, 2018).
2.3.4.2.3 Departure time
Departure time is one of the influential factors on demand for e-scooter-sharing that should be
considered. In Portland, USA, the two peak periods of using e-scooter-sharing are recreational
trips on weekends between 2 and 5 pm and the evening commute on weekdays between 3 and
6 pm (Shaheen and Cohen, 2019). Interestingly, the temporal characteristics of the e-scooter
usage patterns in Minneapolis and Austin are different. More e-scooter traffic in the afternoons
and weekends in Austin, while Minneapolis experienced more evening riding and consistent
daily vehicle miles traveled during the week (Bai and Jiao, 2020). In Austin, USA, the
distribution of hourly e-scooters trip rates on weekdays displays the long afternoon plateau.
Moreover, the average daily use is higher on holidays and weekends. Also important is the
morning distance from the origin to the central business district. It may be because morning
travels are more concentrated around Austin's core (Caspi et al., 2020). In Washington, D.C.,
e-scooter usage varies between weekends and weekdays, while the AM/PM difference is
negligible (Younes et al., 2020). In Indianapolis, USA, most scooter activities are observed
between 11:00 am and 9:00 pm. It is significantly different from the usual AM / PM traffic
peaks. Besides, the use of scooters in the morning was relatively low. This shows that scooters

77
were not a practical option for commuting in the morning to work in Indianapolis (Mathew et
al., 2019).
2.3.4.2.4 Trip purpose
Trip purpose can be influential in choosing e-scooter-sharing. In Oslo, Norway, the two
primary trip purposes of e-scooter users are leisure (40%) and travel to/from school or work
(29%) (Berge, 2019). In Portland, USA, 71% of respondents in a survey reported using e-
scooter-sharing to reach their destination, while 29% chose it for recreational purposes
(Shaheen and Cohen, 2019). In Washington, DC, e-scooter travels originated predominantly
from the public/recreational area and ended in the same land use, while bike-share travels are
primarily home-based commutes (McKenzie, 2019). In Louisville, USA, e-scooters are
probably not being used for commute trips but could be chosen for short commutes (Noland,
2019). Hence, commuting does not seem to be the primary travel purpose. Also, e-scooters
might be an alternative to non-working short trips (Caspi et al., 2020).
2.3.4.2.5 COVID-19
Some studies have shown that the covid-19 epidemic has negatively affected the e-scooter
market (Button et al., 2020). Popov and Ravi (2020) state that COVID-19 is essential to e-
scooter ownership advantages. Also, e-scooter-sharing services are disappearing in more and
more cities as the coronavirus continues to spread worldwide. The simple reason not to use e-
scooter-sharing services is that no one wants to trip by touching brakes and handlebars that
may be infected because many people use the shared e-scooters. Thus, it can have serious
negative consequences for the shared mobility sector. In contrast, Elhenawy et al. (2020) stated
that during the COVID-19 epidemic, more individuals switched to micro-mobility ride-sharing
systems. It is also less affected by COVID-19 than other public transport modes, such as trains
and buses. Hence, COVID-19 effects on the usage of shared e-scooters are controversial.
2.3.4.3 Scooter-sharing characteristics
One of the most important factors influencing e-scooter-sharing usage is the scooter-sharing
characteristics, such as travel comfort, transportation facilities, service quality, and travel cost.
The impact of the main scooter-sharing characteristics is examined in the following.
2.3.4.3.1 Travel cost
E-scooter-sharing subscribers seem to be more sensitive to changes in gasoline prices than
station-based bike-sharing subscribers (Younes et al., 2020). Perceived price is an important
factor in service loyalty, and, in a way, perceived price performance increases service loyalty
through customer satisfaction. It may be especially important for companies offering e-scooter-
sharing services because they are a relatively immature market prone to price fluctuations. E-
scooter-sharing services such as Lime or Bird have recently raised their prices, leading to less
demand for the service. It is also noteworthy that as prices increase, the e-scooter-sharing
service becomes less attractive to users, and users prefer to purchase their e-scooter (Popov and
Ravi, 2020). US cities embrace e-scooters-sharing warmly. Because the price of e-scooters-
sharing is flexible; hence, it is much cheaper than station-based bike-sharing for short-distance
trips (Bai and Jiao, 2020).

78
2.3.4.3.2 Travel comfort
When using e-scooter-sharing, the need for comfort in daily travel is somewhat less than the
need for freedom and time savings. E-scooter-sharing is sensitive to rough roads and pavements
and requires constant attention while riding (Berge, 2019). E-scooter-sharing is a no-sweat way
of reaching your destination. However, there is nowhere to stow groceries or other belongings
for e-scooter users (Schellong et al., 2019). However, in comparison to e-bikes, e-scooter-
sharing has more advantages. The user can stand, which means no wrinkles on clothes for
office workers. In addition, the posture is more comfortable for females who wear dresses or
skirts. Also, unlike bikes, in some places, the e-scooter is not subject to wear a helmet (Sipe
and Pojani, 2018).
2.3.4.3.3 Transportation facilities
The transportation facilities factor is another noticeable element to consider. Interestingly, the
use of e-scooter has a positive relationship with transportation facilities in Austin, USA, but a
negative relationship with the Minneapolis transport facilities. In Austin, people could connect
their transit trips with e-scooters, while in Minneapolis, e-scooters were probably independent
of transportation (Bai and Jiao, 2020). In Austin, USA, e-scooters usage is associated with
areas with bike infrastructure. Also, the origins and destinations of e-scooter trips are associated
with bus stop locations; hence, users may link bus trips and e-scooters (Caspi et al., 2020).
Also, if the streets are equipped with bicycle lanes, they will probably attract more e-scooter
traffic (Zou et al., 2020).
2.3.4.3.4 Service quality
The quality of services is indirectly an important factor in service loyalty. Improving the quality
of the e-scooter-sharing service can also raise customer satisfaction, which is a critical factor
in service loyalty (Popov and Ravi, 2020).
2.3.4.4 Built environment and land use
Land use and accessibility factors are considerable elements impacting the demand for e-
scooter-sharing.
2.3.4.4.1 Land use
The land-use factor is a significant element affecting e-scooter-sharing use. The effect of the
degree of land use mix on e-scooter-sharing use is more significant than the effect of the
percentage of the education level and open space to ride (Jiao and Bai, 2020). In Indianapolis,
the USA, 15% of scooters were used for more than an hour daily. Therefore, it is important to
understand the proportion of scooter use, especially in densely populated areas (Mathew et al.,
2019). In Washington, DC, local arteries and streets with heavy traffic are the most popular
facilities used by the shared e-scooters. It is important to underscore that e-scooter-sharing is
the best solution for high-density downtown (Katona and Juhasz, 2020). In Austin, areas with
higher population density are associated with more e-scooter travel. Also, a shorter distance
from the city center and more complex land use raised the usage of e-scooter (Jiao and Bai,
2020).
Minneapolis and Austin cities differ in size and density but are similar in terms of urban
shape and land-use layout near the city center. The densest use of e-scooters occurred in city

79
university campuses and downtown areas. Also, proximity to the city center and greater land-
use diversity positively correlate with higher e-scooter usage rates in Austin and Minneapolis,
USA. Besides, compared to single-family residential zones, office and institutional land use
are more likely to be associated with higher e-scooter ride rates in both cities. Curiously, e-
scooter-sharing use has a statistically positive relationship with parks and commercial areas
only in Austin (Bai and Jiao, 2020). In Austin, USA, the e-scooter trip is less likely to start and
end in recreational areas and is more likely to do so in industrial, commercial, and residential
areas. Also, in Austin's center, individuals use e-scooter-sharing regardless of the
neighborhood’s wealth; thus, e-scooters are widely used in different areas. Besides, e-scooter-
sharing services can work well on campuses or college towns (Caspi et al., 2020).
2.3.4.4.2 Accessibility
One of the remarkable factors influencing e-scooter-sharing demand is the accessibility factor.
Better access to transit is positively associated with the increased use of e-scooters in Austin
and Minneapolis, USA (Bai and Jiao, 2020). Transit stations and better street connectivity
increase the usage of e-scooter in Austin, USA (Jiao and Bai, 2020). Moreover, increasing
service visibility in popular areas such as bus stops, student residences, and train stations can
alleviate the first-mile / last-mile problems in urban transportation (Popov and Ravi, 2020).
2.3.4.5 Natural Environmental Conditions
The hilliness, weather conditions, and temperature factors are natural environmental conditions
that influence e-scooter-sharing usage.
2.3.4.5.1 Hilliness
E-scooters do not perform well in brick-lined streets or hilly areas (Schellong et al., 2019).
2.3.4.5.2 Weather condition
E-scooters are ill-suited for adverse weather (Schellong et al., 2019). In Louisville, Kentucky,
USA, rain and snow reduced the use of e-scooters. Also, strong winds slightly decreased travel
distance. Travel distance also decreases when it rains (about 0.06 km per cm) (Noland, 2019).
However, compared to station-based bike-sharing users, e-scooter-sharing users are less
sensitive to weather changes (Younes et al., 2020).
2.3.4.5.3 Temperature
Generally, a higher average temperature is unrelated to a higher travel rate. However, it can
lead to faster and longer trips in some places, such as Louisville, Kentucky, USA, where the
average daily minimum and maximum temperatures are -11.67º C and 29.44º C, respectively
(Noland, 2019).
2.3.5 Summary
The socio-demographic characteristics, containing ownership status, occupation and economic
status, age, education level, and gender of users, affect e-scooter usage. Also, trip-related
characteristics, including COVID-19, departure time, travel distance, and trip purpose, affect
e-scooter-sharing usage. Besides, the built environment and land use factors comprising land
use and accessibility factors influence the usage of e-scooter-sharing remarkably. Moreover,

80
natural environmental conditions like hilliness, weather, and temperature affect e-scooter-
sharing demand. Furthermore, scooter-sharing characteristics influence demand, including
service quality, travel cost, travel comfort, and transportation facilities. Table 25 is given to
better view the factors affecting the use of e-scooter-sharing.
Table 25: Influence of factors on the use of e- scooter-sharing.
Factors
Sub-factor
Positive impact
Negative impact
No impact
Reference
Natural
environmental
conditions
Hilliness
-
Brick-lined streets
or hilly areas
-
Schellong et al., 2019
Temperature
-
-
Warm
temperature
(only leads to
faster and
longer trips)
Noland, 2019
Weather
condition
Appropriate weather
Adverse weather,
rain, snow
Wind (only
decreases the
travel distance)
Schellong et al., 2019;
Noland, 2019
Built
environment and
land use
Land use
Proximity to the city
center, greater land-use
diversity, local arteries,
streets with heavy traffic,
parks, high-density
downtown, higher
population density areas,
university campuses,
office and institutional
land use, commercial
areas, more complex
land-use, residential
areas, college towns
-
-
Zou et al., 2020; Katona
and Juhasz, 2020; Bai
and Jiao, 2020; Jiao and
Bai, 2020; Caspi et al.,
2020
Accessibility
Better access to transit,
better street
connectivity, service
visibility
-
-
Bai and Jiao, 2020; Jiao
and Bai, 2020; Popov
and Ravi, 2020
Trip-related
characteristics
Trip purpose
Leisure or recreational
trips travel to/from
school or work, non-
working short trips
Commute trips
-
Berge, 2019; Shaheen
and Cohen, 2019;
Noland, 2019; Caspi et
al., 2020
Trip distance
Short distance, between
half-mile (about 800 m)
and two miles (around
3.22 km)
Long-distance,
more than three
miles
(approximately
4.83 km)
-
Smith and
Schwieterman, 2018;
Schellong et al., 2019;
Todd et al., 2019;
Wüster et al., 2020;
Gössling, 2020
Departure time
Recreational trips on
weekends, evening
commute on weekdays,
holidays, and weekends
Morning
-
Mathew et al., 2019;
Shaheen and Cohen,
2019; Younes et al.,
2020; Bai and Jiao,
2020; Caspi et al., 2020
Covid-19
Covid-19
Elhenawy et al., 2020
Covid-19
Button et al., 2020;
Popov and Ravi, 2020
Travel time
Shorter travel time
-
-
Berge, 2019; Clewlow,
2019; Todd et al., 2019;
Younes et al., 2020
Scooter-sharing
characteristics
Service quality
High-quality
-
-
Popov and Ravi, 2020
Travel cost
Lower price
-
-
Popov and Ravi, 2020;
Bai and Jiao, 2020;
Younes et al., 2020
Travel comfort
No need to wear a helmet
in some places, also
traveling with no
problem with wrinkles,
sweating, or wearing a
skirt
Nowhere to stow
groceries or other
belongings
-
Sipe and Pojani, 2018;
Schellong et al., 2019;
Berge, 2019
Transportation
facilities
Bike-sharing path
-
-
Caspi et al., 2020; Bai
and Jiao, 2020; Zou et
al., 2020

81
Factors
Sub-factor
Positive impact
Negative impact
No impact
Reference
Socio-
demographic
characteristics
Occupation and
economic status
Low-income people,
employed people,
student
-
-
Shaheen and Cohen
(2019); Clewlow, 2019;
Jiao and Bai, 2020;
Caspi et al., 2020
Gender
Males
-
-
Haworth and Schramm,
2019; Berge, 2019; Laa
and Leth, 2020; Curl
and Fitt, 2020
Ownership status
Non-ownership
Ownership
-
Popov and Ravi; Turoń
and Czech, 2019;
Shaheen and Cohen,
2019; Popov and Ravi,
2020
Education level
Well-educated people
-
-
Jiao and Bai, 2020; Laa
and Leth, 2020
Age
Young adult
Elder
-
Haworth and Schramm,
2019; Berge, 2019;
Clewlow, 2019;
Shaheen and Cohen,
2019; Popov and Ravi,
2020; Bai and Jiao,
2020; Laa and Leth,
2020; Curl and Fitt,
2020; Rahimuddin et
al., 2020
Over the past few years, e-scooter-sharing has blossomed as a micro-mobility system that
can alleviate some of the challenges facing today's large cities and pave the way for sustainable
urban transportation development. This study aims to offer a framework that determines the
factors influencing the demand for e- scooter-sharing. These results enable decision-makers or
planners to understand the key elements affecting e-scooter-sharing demand.
This study's key conclusions, separately considering the six factors, are reported in the
following lists.
The most significant socio-demographic characteristics that impact the demand for e-
scooter-sharing are as follows:
• E-scooters cater to many young urban dwellers' special preferences due to youths'
lifestyles and priorities.
• E-scooter-sharing users are primarily male.
• Well-educated people are more interested in using e-scooters.
• The higher the employment rate in the area, the higher the use of e-scooter-sharing.
• E-scooters can be more popular with low-income groups and can potentially help cities
achieve justice goals.
• The lower user's income, the more departures, and arrivals are made on weekday
mornings.
• E-scooters are used by students who are likely to have lower incomes but are not socio-
economically low.
• Not owning an e-scooter has benefits such as shared social benefits and sustainability.
• Elders cannot simply use e-scooters.
• Females may feel more secure when using e-scooters, and e-scooters enable females to
travel long distances.

82
• Although the benefits of owning an e-scooter may undermine loyalty to the e-scooter-
sharing service, the impact is not significant.
The most significant trip-related characteristics that affect the use of e-scooter-sharing are
as follows.
• E-scooter-sharing is chosen chiefly for weekend recreational trips, weekday commutes,
and holidays.
• Using an e-scooter on daily trips, especially compared to bicycles and e-bikes, can save
time.
• Passengers traveling half a mile (about 800 m) and two miles (around 3.22 km) receive
the most out of e-scooters.
• Most travelers who travel more than three miles (approximately 4.83 km) use scooters
to access bus and train stations.
• E-scooters may be an alternative to some non-working short trips, and commuting does
not seem to be the primary trip purpose.
• The use of e-scooters in the morning is relatively low, indicating that e-scooters are not
a suitable transportation option for morning commuting.
• COVID-19 effects on the usage of shared e-scooters are controversial.
The most important scooter-sharing characteristics that affect the use of e-scooter-sharing
are as follows:
• The use of e-scooter-sharing can be increased when the origin and destination of e-
scooter trips are linked to bus station locations or the streets are equipped with bicycle
lanes.
• Higher service quality leads to higher service loyalty.
• When the travel cost of e-scooter-sharing is less than that of bike-sharing, e-scooter-
sharing can attract more people.
• E-scooter-sharing is a no-sweat way of reaching your destination; however, it does not
have any place to stow belongings.
• Riding e-scooter-sharing does not cause wrinkles in clothes. Also, females can easily
ride it with a skirt. Also, in some places, no need to wear a helmet. Hence, the travel
comfort of e-scooter-sharing can be greater than the e-bike.
• For e-scooter-sharing users, the need for comfort in daily travel is somewhat less than
the need for freedom and time savings.
The most remarkable built environment and land use features that influence e-scooter-
sharing use are as follows:
• Proximity to the city center, more complex land-use, greater land-use diversity, local
arteries, streets with heavy traffic, parks, high-density downtown, higher population
density areas, university campuses, college towns, commercial areas, residential areas,
office, and institutional land use can lead to increasing e-scooter use.

83
• Better access to transit is positively associated with the increased use of e-scooter-
sharing.
• Increasing service visibility in popular areas can reduce urban transportation's first/last
mile problems.
The influence of natural environmental conditions on e-scooter-sharing use is as follows:
• E-scooters do not perform well in brick-lined streets or hilly areas.
• E-scooters are not suitable for adverse weather.
• Although higher average temperatures are not associated with higher travel rates, they
can lead to faster and longer trips.
The impact of factors that can affect the demand for e-scooter-sharing is a significant issue
for study. Much research needs to be conducted since not much time has passed since the
emergence of e-scooter-sharing. Furthermore, in most studies, only one or two main factors
affecting the demand for shared e-scooters have been investigated. Further, quantitative
research has considered several factors simultaneously. Therefore, in future research, more
factors should be considered concurrently.
2.4 Definition of the criteria and sub-criteria that impact the
demand for different shared mobility services
A literature review helps determine important criteria and sub-criteria for comparing shared
mobility services, including bike-sharing, car-sharing, and scooter-sharing. Sub-criteria
included in each criterion share some common characteristics among them. Based on the above
literature and knowledge of the author, each criterion includes some sub-criteria, listed in Table
26. Table 26 summarizes the criteria and sub-criteria significantly impacting demand for car-
sharing, bike-sharing, and scooter-sharing services.
Table 26: Criteria and sub-criteria influencing the use of each shared mobility system.
Criteria
Sub-criteria
Shared mobility systems
Car-sharing
Bike-sharing
E-scooter-sharing
Trip-related
characteristics
Travel time
Cervero, 2003; Catalano et al.,
2008; Efthymiou et al., 2013; Kim
et al., 2017c; Carroll et al., 2017
Jensen et al., 2010; Buehler and
Hamre, 2014; Mateo-Babiano et
al., 2016
Berge, 2019; Clewlow,
2019; Todd et al.,
2019; Younes et al.,
2020
Travel distance
Martin and Shaheen, 2011a;
Costain et al., 2012; Wang et al.,
2012; Martínez et al., 2017; Li,
2019; Li, 2019
Fishman, 2016; Campbell et al.,
2016; El-Assi et al., 2017; Li,
2019; Li et al., 2019; Ji et al.,
2020; Chen et al. (2020)
Smith and
Schwieterman, 2018;
Schellong et al., 2019;
Todd et al., 2019;
Wüster et al., 2020;
Gössling, 2020
Departure time
Cervero, 2003; Costain et al.,
2012; Ceccato, 2020
Froehlich et al., 2008;
Kaltenbrunner et al., 2010; Kim
et al., 2012; Faghih-Imani et al.,
2014; O’Brien et al., 2014;
Corcoran et al., 2014; Reiss and
Bogenberger, 2016; Ahillen et
al., 2016; Mateo-Babiano et al.,
2016; Kutela and Kidando, 2017;
Kim et al., 2018; Heaney et al.,
2019; Li et al., 2019; Ji et al.,
2020; Lin et al., 2020
Mathew et al., 2019;
Shaheen and Cohen,
2019; Younes et al.,
2020; Bai and Jiao,
2020; Caspi et al.,
2020
Trip purpose
Cervero, 2003; Martin and
Shaheen, 2011a; Le Vine,
Fishman, 2016; Li and
Kamargianni, 2018; Li, 2019;
Berge, 2019; Shaheen
and Cohen, 2019;

84
Criteria
Sub-criteria
Shared mobility systems
Car-sharing
Bike-sharing
E-scooter-sharing
Adamou, and Polak, 2014; Kim et
al., 2015; Cartenì et al., 2016;
Wang et al., 2017; Le Vine and
Polak, 2019; Jin et al., 2020
Noland et al., 2019; Li et al.,
2019; Chen et al., 2020
Noland, 2019; Caspi et
al., 2020
Travel mode
characteristics
Travel cost
Catalano et al., 2008; Shaheen and
Martin, 2010; Lamberton and
Rose, 2012; De Luca and Di Pace,
2015; Cartenì et al., 2016; Yoon et
al., 2017; Carroll et al., 2017;
Rotaris et al., 2019
Goodman and Cheshire, 2014;
Fishman, 2016; Nikitas, 2018; Li
and Kamargianni, 2018
Popov and Ravi, 2020;
Bai and Jiao, 2020;
Younes et al., 2020
Travel comfort
Schaefers, 2013
Zanotto, 2014; Leister et al.,
2018
Sipe and Pojani, 2018;
Schellong et al., 2019;
Berge, 2019
Infrastructure,
trip end, and en-
route facilities,
transportation
facilities
Irrelevant
Krykewycz et al., 2010; Buck
and Buehler, 2012; Bachand-
Marleau et al., 2012; Rixey,
2013; Croci and Rossi, 2014;
Zhou, 2015; Fishman et al.,
2015; Faghih-Imani and Eluru,
2016b; Mateo-Babiano et al.,
2016; Noland et al., 2016; Jain et
al., 2018; Wang et al., 2018;
Kabak et al., 2018; Wang and
Lindsey, 2019; Xu and Chow,
2019; Duran-Rodas et al., 2019;
Wang and Akar, 2019; Zhao et
al., 2019; Noland et al., 2019
Caspi et al., 2020; Bai
and Jiao, 2020; Zou et
al., 2020
Service quality
Research gap
Research gap
Popov and Ravi, 2020
Helmet
provision
Irrelevant
Bonyun et al., 2012; Kraemer et
al., 2012; Grenier, 2013;
Fishman et al., 2013; Fishman et
al., 2014; Basch and Zagnit,
2014; Basch et al., 2014;
Fishman, 2016
Research gap
Availability
and
accessibility
Land-use
Cervero, 2003; Shaheen and
Rodier, 2005; Millard-Ball, 2005;
Burkhardt and Millard-Ball, 2006;
Kortum and Machemehl, 2012;
Habib et al., 2012; Kopp et al.,
2015; Wagner et al.,2016; Juschten
et al., 2017; Becker et al., 2017a;
Dias et al., 2017; Namazu et al.,
2018; Hu et al., 2018; Ceccato and
Diana, 2021
Buck and Buehler, 2012; Kim et
al., 2012; Hampshire and Marla,
2012; Bachand-Marleau et al.,
2012; Noland et al., 2016; Croci
and Rossi, 2014; Etienne and
Latifa, 2014; Noland et al., 2016;
Zhang, 2017; El-Assi et al.,
2017; Kutela and Kidando, 2017;
Jain et al., 2018; Duran-Rodas et
al., 2019; Wang and Lindsey,
2019; Zhao et al., 2019; Duran-
Rodas et al., 2019; Lin et al.,
2020; Ji et al., 2020
Zou et al., 2020;
Katona and Juhasz,
2020; Bai and Jiao,
2020; Jiao and Bai,
2020; Caspi et al.,
2020
Accessibility
Brook, 2004; Catalano et al., 2008;
Stillwater et al., 2008; Stillwater et
al., 2008; Zheng et al., 2009;
Costain et al., 2012; Kim et al.,
2017b; Juschten et al., 2017
Bachand-Marleau et al., 2012;
Kim et al., 2012; Wang and
Lindsey (2019); Zhao et al.,
2019; Ji et al., 2020
Bai and Jiao, 2020;
Jiao and Bai, 2020;
Popov and Ravi, 2020
Size and age of
stations
Stillwater et al., 2008; Habib et al.,
2012; De Lorimier and El-
Geneidy, 2013
Bachand-Marleau et al., 2012;
Faghih-Imani and Eluru, 2016b;
Wang and Lindsey, 2019; Xu
and Chow, 2019; Wang and
Akar, 2019
Research gap
Natural
environmental
conditions
Hilliness
Irrelevant
Jennings, 2011; Frade and
Ribeiro, 2014; Fricker and Gast,
2016; Bordagaray et al., 2016
Schellong et al., 2019
Weather
condition
Irrelevant
Miranda-Moreno and Nosal,
2011; Gebhart and
Noland, 2014; Corcoran et al.,
2014; Faghih-Imani and Eluru,
2016a; Caulfield et al., 2017;
Fournier et al., (2017); De
Chardon et al., 2017; Martinez,
2017; Kutela and Kidando, 2017;
Sun et al., 2018; Lin et al., 2020;
Eren and Uz, 2020
Schellong et al., 2019;
Noland, 2019
Temperature
Irrelevant
Miranda-Moreno and Nosal,
2011; Faghih-Imani et al., 2014;
Noland, 2019

85
Criteria
Sub-criteria
Shared mobility systems
Car-sharing
Bike-sharing
E-scooter-sharing
Corcoran et al., 2014; Gebhart
and
Noland, 2014; Jing and Zhao,
2015; Faghih-Imani and Eluru,
2016a; El-Assi et al., 2017;
Martinez, 2017; Wang et al.,
2018; Hyland et al., 2018; Kim,
2018; Heaney et al., 2019; Eren
and Uz, 2020; Lin et al., 2020
Air pollution
Irrelevant
Li and Kamargianni, 2018
Research gap
Seasonal effect
Irrelevant
Rudloff and Lackner, 2014;
Godavarthy and Taleqani, 2017;
Kutela and Kidando, 2017;
Heaney et al., 2019; Eren and Uz,
2020
Research gap
Socio-
demographic
characteristics
Gender
Cervero, 2003; Martin and
Shaheen, 2011a; Firnkorn and
Müller, 2012; Morency et al.,
2012; Kawgan-Kagan, 2015; Kopp
et al., 2015; Ciari et al., 2015;
Wielinski et al., 2015; Cartenì et
al., 2016; Becker et al., 2017a;
Yoon et al., 2017; Ceccato and
Diana, 2021; Hu et al., 2018;
Shaheen et al., 2018; Acheampong
and Siiba, 2020
Ogilvie and Goodman, 2012;
Zanotto, 2014; Vogel et al.,
2014; Goodman and Cheshire,
2014; Buehler and Hamre, 2014;
Ricci, 2015; Faghih-Imani and
Eluru, 2015; Fishman et al.,
2015; Fishman, 2016; Raux et
al., 2017; Nikitas, 2018; Wang
and Akar, 2019; Li et al., 2019;
Chen et al., 2020
Haworth and
Schramm, 2019;
Berge, 2019; Laa and
Leth, 2020; Curl and
Fitt, 2020
Age
Brook, 2004; Lane, 2005; Millard-
Ball, 2005; Burkhardt and Millard-
Ball, 2006; Shaheen and Martin,
2010; Martin et al., 2010; Martin
and Shaheen, 2011a; Firnkorn and
Müller, 2012; Kortum and
Machemehl, 2012; Sioui et al.,
2013; Wielinski et al., 2015; Kim
et al., 2015; Cartenì et al., 2016;
Clewlow, 2016; Becker et al.,
2017a; Perboli et al., 2017;
Ceccato and Diana, 2021; Vinayak
et al., 2018; Shaheen et al., 2018;
Rotaris and Danielis, 2018;
Burghard and Dütschke, 2019;
Ceccato, 2020
Fuller et al., 2011; Vogel et al.,
2014; Zanotto, 2014; Fishman et
al., 2015; Ricci, 2015; Raux et
al., 2017; Nikitas, 2018; Wing et
al., 2018; Li et al., 2019; Eren
and Uz, 2020; Chen et al., 2020
Haworth and
Schramm, 2019;
Berge, 2019; Clewlow,
2019; Shaheen and
Cohen, 2019; Popov
and Ravi, 2020; Bai
and Jiao, 2020; Laa
and Leth, 2020; Curl
and Fitt, 2020;
Rahimuddin et al.,
2020
Occupation
and economic
status
Millard-Ball, 2005; Cervero et al.,
2007; Martin et al., 2010; Martin
and Shaheen, 2011a; Efthymiou et
al., 2013; Efthymiou and
Antoniou, 2014; Kim et al., 2015;
Kawgan-Kagan, 2015; Clewlow,
2016; Dias et al., 2017; Yoon et al.,
2017; Winter et al., 2017; Ceccato
and Diana, 2021; Vinayak et al.,
2018; Shaheen et al., 2018;
Ceccato, 2020
Maurer, 2011; Rixey, 2013;
Zanotto, 2014; Fishman et al.,
2015; Ricci, 2015; Murphy and
Usher, 2015; Raux et al., 2017;
Li et al., 2019
Shaheen and Cohen
(2019); Clewlow,
2019; Jiao and Bai,
2020; Caspi et al.,
2020
Vehicle
ownership
status
Cervero and Tsai, 2004; Cervero et
al., 2007; Celsor and Millard-Ball,
2007; Martin et al., 2010; Martin
and Shaheen, 2011a; Ter Schure et
al., 2012; Habib et al., 2012; Sioui
et al., 2013; Efthymiou and
Antoniou, 2014; De Luca and Di
Pace, 2015; Mishra et al., 2015;
Kopp et al., 2015; Clewlow, 2016;
Juschten et al., 2017; Dias et al.,
2017; Ceccato and Diana, 2021;
Namazu et al., 2018; Burghard and
Dütschke, 2019; Mishra et al.,
2019; Lempert et al., 2019;
Ceccato, 2020
Bachand-Marleau et al., 2012;
Shaheen and Guzman, 2011;
Fishman et al., 2015; Chen et al.,
2020
Popov and Ravi; Turoń
and Czech, 2019;
Shaheen and Cohen,
2019; Popov and Ravi,
2020
Household size
Cooper et al., 2000; Millard-Ball,
2005; Kortum and Machemehl,
2012; Ceccato and Diana, 2021;
Ceccato, 2020
Research gap
Research gap

86
Criteria
Sub-criteria
Shared mobility systems
Car-sharing
Bike-sharing
E-scooter-sharing
Marital status
Celsor and Millard-Ball, 2007;
Efthymiou and Antoniou, 2014;
Carroll et al., 2017
Research gap
Research gap
Presence of
children
Sioui et al., 2013; Coll et al., 2014;
Kopp et al., 2015; Carroll et al.,
2017; Kim et al., 2017; Dias et al.,
2017; Vinayak et al., 2018; Rotaris
and Danielis, 2018
Research gap
Research gap
Residence
status
(permanent
residence or
not)
Research gap
Li et al., 2019; Du et al., 2019
Research gap
Education level
Cooper et al., 2000; Brook, 2004;
Millard-Ball, 2005; Burkhardt and
Millard-Ball, 2006; Martin et al.,
2010; Shaheen and Martin, 2010;
Martin and Shaheen, 2011a;
Firnkorn and Müller, 2012; Wang
et al., 2012; Coll et al., 2014;
Kawgan-Kagan, 2015; Kopp et al.,
2015; Dias et al., 2017; Becker et
al., 2017a; Juschten et al., 2017;
Carroll et al., 2017; Prieto et al.,
2017; Shaheen et al., 2018;
Vinayak et al., 2018; Ceccato,
2020
Fuller et al., 2011; Bachand-
Marleau et al., 2012; Zanotto,
2014; Fishman et al., 2015;
Ricci, 2015; Li et al., 2019;
Cheng et al., 2020
Jiao and Bai, 2020;
Laa and Leth, 2020

87
Chapter 3
Methodology: Multi-Criteria Decision-
Making Methods
After defining the research questions in Chapter 1 and providing a comprehensive literature
view to understand the important factors affecting each shared mobility service (car-sharing,
bike-sharing, and scooter-sharing), in this section, the proper method according to the purpose
of this study should be chosen. This chapter explains the different methods of Multiple-Criteria
Decision-Making (MCDM). MCDM is also known as Multiple-Criteria Analysis (MCA) or
Multiple-Criteria Decision Analysis (MCDA). In this research, MCDM is used for greater
clarity.
Decision-making is commonly described as the cognitive process of choosing an
alternative from a set of alternatives. In the MCDM problem, the decision-maker has to identify
the best alternative from a set of alternatives taking into account a set of criteria.
A discrete MCDM problem is usually indicated as a matrix, as presented in Eq. (1)
(Kalpoe, 2020b).
  



   
   
   
   
(1)
Where,
󰇝󰇞: a set of alternatives
󰇝󰇞: a set of criteria
: the score (indicator value) of alternative  (=1,…,) concerning criterion  (=1,…,)

88
Choosing the best (e.g., most favorable, most substantial) alternative (with the best
value) is the purpose of the MCDM problem, as displayed in Eq. (2) (Jong and Stone, 1976).
The highest  represents the most desirable alternative. Hence,  is the overall value of
alternative  that can be computed utilizing the additive value function as shown in Eq. (2).
When the weight  is assigned to criterion ,  is determined by multiplying the score  with
the respective weight  of criterion . Hence, a set of alternatives and a set of decision criteria
by which the alternatives can be evaluated are required. The weight of the criteria is then
determined, and there are different methods for inferring the weight of the criteria in the
literature.





(2)
Where
: overall value of alternative 
: weight assigned to criterion 

 : the normalization of each score (indicator value) of alternative (), (=1,…,, and
=1,…,)
Scores are gathered from accessible sources of data (for the objective and accessible
ones like the price) or measured utilizing qualitative methods such as the Likert scale or
computed like the criteria weights (for the subjective ones such as quality) and normalized
utilizing a normalization formula. Hence, to normalize, if the alternative scores (performance
matrix) are in different scales, the scores have to be normalized, as mentioned in Eq. (3)
(Brispat, 2017). In Eq. (3), 
 is the normalization of each score of alternative (), which
can be determined by dividing each score of alternative () by the largest value of that score
among the alternatives 󰇝󰇞󰇜. The inverse equation is applied for a criterion value, such as
price, considered a negative value.








󰇝󰇞󰇛󰇜
 
󰇥󰇦󰇛󰇜
(3)
: the score of alternative  concerning criterion 

 : the normalization of each score of alternative ()
In this study, Perception-Based Analysis (PBA) is conducted. Different stakeholders
participate in perception-based analysis, and quantitative and qualitative data can be considered
to specify different stakeholders' perceptions of shared mobility systems (Scholten et al., 2017).
It assists in calculating stakeholders' perceptions, including their opinion, interpretations, and
understanding. As a result, it can provide more insights into the stakeholders' perceptions,

89
leading to a clearer decision-making process. In this regard, information regarding stakeholder
perceptions can be utilized when determining the problem, and possible solutions to deal with
it are provided. Besides, after selecting, confirming, and implementing the system, information
based on perception can be utilized to raise the level of satisfaction (e.g., user satisfaction),
which can even change the perception of users (Brispat, 2017).
3.1 Multi-actor multi-criteria analysis
Stakeholders are an important aspect to consider. Therefore, how different stakeholders rate
the importance of comparison factors must be determined. Hence, it is necessary to specify the
appropriate method for the analysis, considering the various stakeholders. In this regard, prof.
Cathy Macharis developed Multi-Actor Multi-Criteria Analysis (MAMCA) in 2005. This
method can be described as a multi-criteria decision analysis that enables decision-makers to
evaluate different projects simultaneously (Macharis et al., 2010). One of the most important
advantages of MAMCA is that MAMCA explicitly considers the views of different
stakeholders. It is important to decide which investment in shared mobility will be most
efficient. Stakeholder participation in the early stages gives policymakers an understanding of
their problem. It also allows them to understand the views of other stakeholders. Figure 2
indicates the seven steps required to perform a MAMCA, as defined by Macharis et al. (2010).
Figure 2: Various steps of the MAMCA method (Macharis et al., 2010).
The steps presented in Figure 2 for the MAMCA method can be described as follows (Macharis
et al., 2012).

90
3.1.1. Defining the problem and specifying alternatives (step 1)
In the first step, the problem must be defined, and several possible alternatives must be
specified. These alternatives can be evaluated later.
3.1.2. Stakeholder analysis (step 2)
This stage is defined as the stakeholder analysis in which all important stakeholders are
identified because considering important stakeholders in the early stages will benefit the result.
Analyzing stakeholders reveals certain aspects, such as priorities, problems, interests, and
conflicts, in the early stages of decision-making. This can be further considered in the overall
process and lead to an improvement in the final result. In addition, this analysis also provides
insight into the project policy level, which clarifies the impact of the project, and the
governmental level (municipal, provincial, national, European) can be considered (if needed in
the study).
3.1.3. Specify criteria and weights (step 3)
The third step is to define the criteria for stakeholders and set weights to indicate their
importance. The criteria are selected based on the stakeholders’ objectives and the purpose of
the considered alternatives. It also means that the different sets of criteria can be important for
each stakeholder group based on their specific goals. In order to show the stakeholders involved
with their goals and objectives, it is possible to provide a hierarchical criteria tree (at this stage).
With the stakeholders, weights can be determined based on the amount of value assigned to
their objectives. These weights then show the importance of the criteria. Finally, if necessary,
it is possible to assign weight to stakeholders. These can show the importance of stakeholders
in the decision-making process.
3.1.4. Criteria, indicators, and measurement methods (step 4)
In the fourth step, the indicators are specified for the criteria set in step 3. The previously
specified stakeholder criteria are ‘operationalized’ by constructing indicators (also called
variables or metrics) that can be applied to gauge whether an alternative contributes to each
metric or to what extent.
3.1.5. Overall analysis and ranking (step 5)
At this stage, each alternative is evaluated and compared using the criteria and indicators
mentioned above. This allows further elaboration on the alternatives in a way that translates to
the scenarios. Once the scenarios are identified, an evaluation table can be provided for each
stakeholder.

91
3.1.6. Results (step 6)
After a general analysis and ranking, the proposed alternative classification can be provided.
This step helps decision-makers in their decision-making process by pointing out which criteria
have a positive or negative impact on alternatives for each stakeholder. This determines the
preference of each stakeholder for each alternative and the importance of the alternative for
each stakeholder.
3.1.7. Implementation (step 7)
Finally, the information and data collected can formulate a policy recommendation for the
decision-makers. Macharis et al. (2012) outlined two implementation approaches from the
decision-makers perspective. The first approach is implementing the alternative that benefits
society the most. The second approach is an alternative implementation that helps to consider
all stakeholders' interests and make compromises.
3.2 Presentation of different MCDM methods
One of the appropriate methods for performing PBA is MAMCA. In the third step of the
MAMCA, the weight of the criteria must be well determined to calculate each stakeholder's
perception. To do this, different MCDM methods can be combined with MAMCA. To find the
most suitable method to combine it with MAMCA, different MCDM methods should be
identified in this study. A comparison between them is essential to find the best method. This
chapter is a way to understand which MCDM method is suitable for combining with MAMCA
to conduct PBA and why.
It is important to note that although there are various MCDM methods in the literature,
the following MCDM methods are chosen for comparison in this study. This is because they
are broadly used in the literature (Triantaphyllou, 2000; Mulliner et al., 2016; Kolios et al.,
2016; Serrai et al., 2017). In this regard, it can be mentioned that Yannis et al. (2020) identified
the most commonly used MCDM techniques in the transport sector. It was figured out that
almost 29% of the studies in the transportation field applied the AHP method. Besides, each of
the following three methods was used in 10% of studies: Elimination and Choice Translating
Reality, Preference ranking organization method for enrichment evaluation, and the Weighted
Product Model. The Technique for Order of Preference by Similarity to Ideal Solution (6%)
and MAMCA (6%) are other important MCDM methods. These well-known methods account
for about 71% of the MCDM methods in the literature. Also, Brispat (2017) emphasized the
importance of the following methods among MCDM methods, especially the Best-Worst
Method.
1. Elimination and Choice Translating Reality
2. Weighted Sum Model
3. Weighted Product Model
4. Analytic Hierarchy Process
5. The Technique for Order of Preference by Similarity to Ideal Solution

92
6. Preference ranking organization method for enrichment evaluation
7. Best-Worst Method
After a brief description of all these methods, the decision on the most appropriate MCDM
method for PBA is made in Section 3.3.
3.2.1 Elimination and choice translating reality
The Elimination and Choice Translating Reality (ELECTRE) method was first introduced
around 1966 by Bernard Roy, and it can be described as a pairwise comparison method
(Benayoun et al., 1966). ELECTRE is run by comparing two alternatives for each criterion.
This prevents ELECTRE from always being able to categorize the most interesting option,
which can be an important drawback depending on the purpose of the problem (Triantaphyllou,
2000). However, when a situation with few criteria and a large number of alternatives occurs
(Lootsma, 1990), ELECTRE may be a great choice for comparing different solutions. This
method can also deal with both quantitative and qualitative factors simultaneously. However,
since ELECTRE can be described as a complex decision method, a large amount of data is
needed to perform the proper analysis. This method can be used in different contexts to
determine which alternatives are preferred according to a set of criteria (Vahdani et al., 2010).
To perform ELECTRE analysis, concordance and discordance indices are considered
(Roy, 1990). Comparing alternative  with alternative , the concordance index
demonstrates when the criteria of one alternative prevail over the criteria of another alternative
(). Conversely, the discordance index indicates when the criteria of  predominate
over that of alternative  (). Finally, Eq. (4) estimates the concordance index (Botti
and Peypoch, 2013).
󰇛󰇜󰆓

(4)
Where 󰇛󰇜 is the concordance index, and  and  represent all criteria and the concordance
criteria, respectively.
Eq. (5) calculates the discordance index (Botti and Peypoch, 2013).
󰇛󰇜 
󰇝󰇞
(5)
Where 󰇛󰇜 is discordance index
• : performance of alternative  with criterion .
• : maximum difference in the performance of the alternatives.
3.2.2 Weighted sum model
One of the easiest and most common methods of MCDM is the Weighted Sum Model (WSM)
(Kolios et al., 2016). This method was developed in 1967 by Peter C. Fishburn; it is easy to

93
use and can be utilized in combination with other methods. The WSM method compares
alternatives based on a set of specific criteria. First, each criterion is given a certain weight.
Then, the optimal solution is easily provided by multiplying the weight of the criteria by the
score of the alternatives.
The WSM problem leads to finding the optimal solution for Eq. (6) (Fishburn, 1967).




(6)
Where 

 indicates the weighted sum score obtained by multiplying the weights by the alternative
scores. The is the score of alternative  concerning criterion . The  is the weight of
criterion .
It is essential to mention that one of the disadvantages of WSM is that when it comes
to using qualitative and quantitative comparison factors, it becomes difficult to do so. This
change in the optimal solution can also occur when some scores are exaggerated.
3.2.3 Weighted product model
The weighted Product Model (WPM) method is an MCDM method with many similarities to
the above-introduced WSM (Kolios et al., 2016) and was developed in 1969. However, the
most significant difference with WSM is that a WPM uses multiplication to calculate the
optimal solution instead of the sum (Triantaphyllou, 2000). Eq. (7) shows a comparison
between the alternatives  and . If  is greater than or equal to 1, the alternative  is
preferred over the alternative .
The optimal solution Is found using Eq. (7) (Bridgman, 1922; Miller and Starr, 1963).
󰇧
󰇨󰇛
󰇜


(7)
Where  represents the number of criteria and 
 is a comparison between the alternatives
 and . The  shows the score of alternative  concerning criterion .  is the weight of
criterion .
3.2.4 Analytic hierarchy process
Thomas L. Saaty developed the Analytic Hierarchy Process (AHP) method in 1980. This
method is mainly used in considering conflicting criteria and energy planning (Kolios et al.,
2016). Conflicting criteria are typical in evaluating alternatives. Typical examples of criteria
that conflict with each other are a measure of quality versus price. There is even a case of
developing an AHP-based approach to dealing with problems where uncertain data is available
(Cobuloglu and Büyüktahtakın, 2015). The AHP method used hierarchical structure and
pairwise comparison to decide complex decision-making problems.

94
3.2.4.1 Hierarchical structure (step 1)
The first step involves creating a decision problem in a hierarchical structure. At the top of the
structure is the purpose of decision-making. In addition, the criteria and sub-criteria influencing
decision-making are at lower levels. Finally, alternatives are placed at the bottom of the
structure.
3.2.4.2 Criteria weights (step 2)
In the second step, the weight of each criterion must be obtained. The pairwise comparison
matrix () or the judgment matrix must be compiled. Each aspect in the matrix, , can be
defined as the importance of criterion  relative to criterion  by considering the alternative.
Eq. (8) shows the weight vector.
󰇛󰇜
(8)
Where  reflects the importance of the -th criterion and is estimated as the means of the
inputs of row  of the normalized matrix  (Saaty, 1980).
Eq. (9) and Eq. (10) are used to examine the consistency of pairwise comparisons (Saaty, 1980).





(9)
Where  indicates the largest eigenvalue of the Matrix .
After finding the maximum eigenvalue (), the Consistency Index () is defined as
presented in Eq. (10) (Saaty, 1980).
󰇛󰇜
 
(10)
Once the  is found, the Consistency Ratio () in the AHP method can be calculated by
dividing the  by the Random Index () to determine whether the degree of consistency is
satisfactory. To do this,  must be defined.  is the average of  values of various sizes of
comparison matrices. In the literature, different authors have calculated and obtained different
 depending on the simulation method and the number of matrices generated involved in
the process. For example, Lane and Verdini (1989), Golden and Wang (1990), and Noble
(1990) performed 2500, 1000, and 5000 simulation runs. Besides, Forman (1990) provided
values for matrices of sizes 3 through 7 using examples from 17672 to 77487 matrices. Tumala
and Wan (1994) subsequently performed the experiment with 4600 to 470000 matrices.
Furthermore, Saaty (1980) simulated the experiment with 500 matrices with the following
algorithm, shown in Table 27.
The steps of the algorithm were (Saaty, 1980);
• Generate a random matrix (Uniform distribution)
• Calculate the corresponding Cis (for each matrix).
• Obtain the average of these values for each size (RI of each size).

95
Table 27: RI for different values n (Saaty, 1980).

2
3
4
5
6
7
8
9
10

0
0.58
0.90
1.12
1.24
1.32
1.41
1.45
1.49
If 
 , serious inconsistencies may present, while if 
 , the
degree of consistency is considered satisfactory.
3.2.4.3 Performance alternatives for criteria (step 3)
The third step is to find the score of each alternative for each criterion. Finally, after calculating
the score of each criterion, the overall score can be determined in the last step.
3.2.4.4 Alternative ranking (step 4)
In the fourth step, the score of the alternatives, , is calculated according to Eq. (11) (Saaty,
1980).



 


(11)
Where,
: the score of the -th alternative
 is the number of alternatives
 is the number of criteria
 is the weight of importance of the -th criterion.
 represents the actual value of the -th alternative in terms of the -th criterion
3.2.5 Technique for order preference by similarities to ideal solution
Technique for Order Preference by Similarities to Ideal Solution (TOPSIS) method is widely
used in various research fields (Kolios et al., 2016) and Hwang and Yoon developed it in 1981.
This method uses the Euclidean distance to find the best solution at the closest (shortest
distance) possible to the ideal alternative and, at the same time, the farthest (longest distance)
from the most negative solution. Both the best and the most negative solutions are obtained
from this method, and any criterion can change utility (Triantaphyllou, 2000). Finally,
changing the utility for each criterion can lead to an ideal and non-ideal solution and an optimal
alternative in this range. Figure 3 displays the necessary methodological steps.

96
Figure 3: RI for different values n (Saaty, 1980).
The first four steps are similar to the steps in the other methods. An explanation of the
following steps is given below.
3.2.5.1 Positive and negative ideal solutions (step 1)
The positive ideal  and negative ideal solution  are derived as given in Eq. (12) and Eq.
(13), respectively. In these equations, 󰆒 and  are associated with the benefit and cost criteria
(positive and negative variables) (Kolios et al., 2016).
󰇝󰇞󰇥󰇡󰇻󰇢󰇡󰇻󰇢󰇦
(12)
󰇝󰇞󰇝󰇡󰇻󰇢󰇡󰇻󰇢󰇞
(13)
Where,
: the positive ideal
: the negative ideal solution
: normalized decision values
: benefit criteria

97
: negative criteria
3.2.5.2 Relative closeness (step 2)
The n-dimensional Euclidean distance is applied to calculate the distance from the alternatives
to  and .  is calculated in Eq. (14) as the separation of each alternative from the ideal
solution. The separation from the negative ideal solution,  is given in Eq. (15) (Kolios et
al., 2016).
󰇛󰇜

 
(14)
󰇛󰇜

 
(15)
Where,
,  : n-dimensional Euclidean distance
: normalized decision values
, the relative proximity to the ideal solution of each alternative is calculated as shown in Eq.
(16) (Kolios et al., 2016).
 

(16)
Where,
: ideal solution of each alternative
With 1 ≥  ≥ 0, where , if  and , if 
3.2.5.3 Solution ranking (step 3)
After sorting the  values, the maximum value corresponds to the best solution to the problem.
The best alternative should be the shortest distance from A+ and the longest distance from the
non-ideal solution.
3.2.6 Preference ranking organization method for enrichment evaluation
Brans developed the Preference Ranking Organization Method for Enrichment Evaluation
(PROMETHEE) method in 1985 (Brans and Vincke, 1985; Brans et al., 1986) and is widely
applied to problems in the energy sector (Kolios et al., 2016). This method uses pairwise
comparisons to provide an overall ranking of options based on positive and negative prediction
flows. PROMETHEE is an easy-to-use method, especially compared to other MCDM methods

98
(Tuzkaya et al., 2010). In addition, PROMETHEE can deal with quantitative and qualitative
factors (Serrai et al., 2017).
Figure 4 displays the steps of the PROMETHEE method, and below figure 4, an
explanation of the method and its five steps is given (Brans et al., 1986; Geldermann and Rentz,
2001; Cao et al., 2006; Tuzkaya et al., 2010; Vulević and Dragović, 2017).
Figure 4: PROMETHEE methodology (Kolios et al., 2016).
3.2.6.1 Preference function (step 1)
First, each criterion’s preference function and weight have to be specified. In order to
demonstrate the importance of each criterion, a certain weight is given to them. If the decision-
maker thinks that all the criteria are equal, they will be assigned the same weight; they do not
need to be normalized.
3.2.6.2 Comparison between alternatives (step 2)
Eq. (17) estimates the global preference index to specify alternative preference over  and
associated criteria  (Brans and Vincke, 1985).
󰇛󰇜󰇛󰇜
(17)
Where,
(a, b): alternatives
: criterion
󰇛󰇜: the difference between evaluating alternatives a and b on the criterion. 󰇛󰇜
󰇛󰇜󰇛󰇜.
󰇛󰇜: the preference of alternative  with regard to alternative  on each criterion as a
function of 󰇛󰇜.
3.2.6.3 Alternative comparison and criteria matrix (step 3)
Eq. (18) determines the amount of preference between a and b (Brans and Vincke, 1985).

99
󰇛󰇜󰇛󰇜

 
(18)
Where, 󰇛󰇜 of  over  (from 0 to 1) is defined as the weighted sum 󰇛󰇜 for each
criterion, and  is the weight associated with th criteria. 󰇛󰇜shows the preference
function  indicates the weight of the criteria .
3.2.6.4 Partial rankings (step 4)
Eq. (19) and Eq. (20) estimate positive outranking flow (󰇛󰇜󰇜 and negative outranking flow
(incoming flow) (󰇛󰇜󰇜, respectively (Brans and Vincke, 1985).󰇛󰇜 indicates how an
alternative a is superior to the others. This is its power and superior character. The higher
󰇛󰇜, the better the alternative. On the other hand, 󰇛󰇜 shows how an alternative “a” is
outranked by all the others. It is its weakness, its outranked character. The lower 󰇛󰇜, the
better the alternative.
󰇛󰇜 
󰇛󰇜
 
(19)
󰇛󰇜 
󰇛󰇜
 
(20)
3.2.6.5 Final rankings of alternatives (step 5)
Finally, the net outranking flow 󰇛󰇜 for each alternative is measured using Eq. (21) (Brans
and Vincke, 1985).
󰇛󰇜󰇛󰇜󰇛󰇜
(21)
The higher 󰇛󰇜 and the lower 󰇛󰇜 means a more positive alternative.
3.2.7 Best Worst Method
The Best Worst Method (BWM) is a vector-based multi-criteria decision-making method
developed by Jafar Rezaei in 2015. This method can be described as a pairwise comparison
between a set of criteria for determining the weight () of the criteria. Pairwise comparison
 designates how much an individual prefers criterion  to criterion . For determination of
such preference, Likert scales (for example, very low…very high) can be used with the
corresponding numerical scale, such as:
0.1, 0.2, …, 1 (0.1: Equally important, …, 1:  is much more important than ).
1, 2, …, 100 (1: Equally important, …, 100:  is much more important than ).
1, …,9 (1: Equally important, …, 9:  is much more important than ).

100
From the set of criteria, participants choose one criterion they consider the most
important (best) and the least important (worst). The best criterion is then compared to the
remaining one, and the same is done for the worst.
The original BWM is presented as a nonlinear optimization problem (Rezaei, 2015).
There is also a linear approximation (Rezaei, 2016), a multiplicative version (Brunelli and
Rezaei, 2019), group decision-making with the BWM (Mou et al., 2016; Hafezalkotob and
Hafezalkotob, 2017; Mohammadi and Rezaei, 2020), and some hybrid versions like BWM-
MULTIMOORA (Hafezalkotob et al., 2019) and BWM-VIKOR.
This approach is also widely used in many real-world applications containing, but not
limited to, supply chain management (Rezaei et al., 2015; Rezaei et al., 2016; Ahmad et al.,
2017; Ahmadi et al., 2017; Vahidi et al., 2018; Gupta and Barua, 2018; Kusi-Sarpong et al.,
2019), transportation and logistics (Rezaei et al., 2017; Groenendijk et al., 2018; Rezaei et al.,
2019), technology management (Gupta and Barua, 2016), science and research assessment
(Salimi and Rezaei, 2016; Salimi, 2017), risk management (Torabi et al., 2016) and energy
(Gupta, 2018; Ren, 2018). Table 28 lists some of the studies in which BWM is used for various
research areas.
Table 28: Some of the studies that applied BWM.
Type of
study
Application
area
Data
source
Number of
respondent
s
Method
Numbe
r of
criteria
Geographi
c coverage
Useful for
Authors
Applicatio
n
Information
Sharing
Arrangement
s
Interview
4
BWM
16
Internationa
l
comparison
All
Stakeholders
Praditya and
Janssen,
2017
Case study
-
-
-
Group decision-
making method
based on BWM
-
-
-
Safarzadeh
et al., 2018
Case study
Equipment
selection
Secondar
y Use
BWM,
MULTIMOOR
A, weighted
aggregated sum
product
assessment
9
-
-
Hafezalkoto
b et al., 2018
Review
paper
-
-
-
-
-
-
-
Mi et al.,
2019
Case study
Maintenance
evaluation of
hospitals
Interview
-
Fully fuzzy
BWM
8
Metropolita
n Level
-
Karimi et
al., 2020
Case
Study
Introducing
BWM
Ad-hoc
Survey
46
BWM
6
Rezaei,
2015
Case
Study
Introducing
linear BWM
Ad-hoc
Survey
-
Linear BWM
-
-
-
Rezaei,
2016
Applicatio
n
Supply chain
Sustainability
Ad-hoc
Survey
48
BWM
6
Internationa
l
comparison
Stakeholders
, integrated
oil and gas
companies
Sadaghiani
et al., 2015
Case
Study
Companies
Interview
-
BWM
12
-
Companies
Rezaei et al.,
2015
Applicatio
n
Transportatio
n
Ad-hoc
Survey
-
BWM
8
Regional
Level
Dairy
industry
Sharma et
al., 2019
Case
Study
Transportatio
n
Ad-hoc
Survey
7
Rough BWM-
Rough
WASPAS
8
-
-
Stević et al.,
2018
Research
paper
Transportatio
n
Ad-hoc
Survey
19
BWM
17
Internationa
l
comparison
Industry and
policy
Rezaei et al.,
2019

101
Type of
study
Application
area
Data
source
Number of
respondent
s
Method
Numbe
r of
criteria
Geographi
c coverage
Useful for
Authors
Case
Study
Transportatio
n
Ad-hoc
Survey
140
BWM
7
Metropolita
n Level
Government
s and
transport
operators
Groenendijk
et al., 2018
Case
Study
--
-
-
-
-
-
-
Brunelli and
Rezaei,
2019
Case study
-
-
-
-
-
-
-
Mohammad
i and
Rezaei,
2020
Research
paper
Zhang et al.,
2017
Case study
Freight
transportatio
n
Ad-hoc
Survey
50
BWM
6
Internationa
l
comparison
Government
s,
policymaker
s, decision-
makers, and
researchers
Liu, 2016
Case
Study
Transportatio
n
-
-
BWM
3
National
level
Supply
freight
Rezaei et al.,
2017
In order to perform the Best Worst Analysis, the following five steps are necessary, which are
described based on the Rezaei (2015, 2016) papers.
3.2.7.1 Definition of the decision criteria (step 1)
A set of decision criteria must first be determined. If the number of criteria is more than nine,
if possible, they can be classified into different groups because, in general, humans can only
compare seven ± two attributes (Miller and Starr, 1963; Glassman et al., 1994). In that case,
there are main criteria and their sub-criteria. The weights obtained for the sub-criteria of the
BWM are called local weights. The local weights can only be utilized to compare the
importance of sub-criteria belonging to the same main criterion. For each sub-criterion, the
global weight can be acquired by multiplying each local weight of the sub-criterion by the
weight of its respective main-criteria. These weights are called ‘global weights’ because they
can be compared in importance, regardless of the classification (main criteria) to which they
belong.
At this stage, a set of criteria 󰇝󰇞 is selected for decision. These are criteria
that can be compared to determine the best result. The set of decision criteria for different
decision-makers might vary (if needed). For further understanding, Figure 5 shows the set of
criteria from 1 to n.
Figure 5: Set of criteria from 1 to n.
3.2.7.2 Determine the best and the worst criteria (step 2)
The best criterion (e.g., most important, most desirable) and the worst criterion (e.g., least
important, least desirable) must be designated. The decision-maker generally picks the best and

102
worst criteria at this stage, and there is still no comparison. For better insight, Figure 6 displays
the selection criteria for the best and worst.
Figure 6: Choosing the criteria of the best and the worst.
3.2.7.3 Determining preference of best criterion over other criteria (step 3)
The strength of the preference of the best criterion over other criteria is designated utilizing a
number between one and nine (or different scales). The number one meaning is an equal
preference between the best and the other criterion. On the other hand, the number nine means
an extreme preference for the best criterion over another. The result of this stage is the vector
of Best-to-others, which is as follows: 󰇛󰇜, Where  shows the
preference of the best criterion  over criterion , and it can be concluded that . For
more apprehension, Figure 7 presents the preference of the best criterion over other criteria.
Figure 7: The preference of the best criterion over other criteria.
3.2.7.4 Determining preference of other criteria over worst criterion (step 4)
By utilizing a number between one and nine, the preference of all criteria over the worst
criterion is designated. The result of this stage is the vector of others-to-worst, which is as
follows: 󰇛󰇜 , where the  states the preference of criterion 
over the worst criterion ; it can be concluded that . For further comprehension,
Figure 8 demonstrates the preference of all criteria over the worst criterion.

103
Figure 8: The preference of each criterion over the worst criterion (Rezaei, 2015).
3.2.7.5 Finding the optimal weights: the first approach
Optimal weights 󰇛󰇜 must be calculated. The optimal weight of the criteria
meets the following conditions: For each pair of  and ,  and
.
Hence, to achieve these conditions for all j, the maximum value of the set 󰇝


󰇞 should be minimized. The problem can be formulated as indicated in Eq. (22) (Rezaei,
2015).

󰇝

󰇞
Subject to 


(22)
Problem (4) can be converted (converted min-max) to Eq. (23) (Rezaei, 2015).

Subject to 






(23)
For each value of , multiply the first set of the constraints of Eq. (23) by  and the second set
of constraints by , the solution space of Eq. (23) is an intersection of  linear

104
constraints. It includes 󰇛󰇜 comparison constraints and a constraint for the sum of the
weights; hence, the value of  is given large enough that the solution space is not empty.
Optimal weights 󰇛󰇜 and  are obtained by solving Eq. (23).
3.2.7.5.1 Consistency ratio in BWM
A comparison is entirely consistent when , for all j, where the preference of
the best criterion over the criterion j is represented as ,  is the preference of criterion j
over the worst criterion, and the preference of the best criterion over the worst criterion is
indicated as  (Rezaei, 2015). For more understanding, Figure 9 shows the concepts of 
, and .
Figure 9: The concepts of  , and .
3.2.7.5.1.1 Consistency ratio definition in BWM (output-based approach)
Since there is probably no full consistency, the level of consistency can be calculated utilizing
a strong indicator called the Consistency Ratio (CR). Calculating the minimum consistency of
comparison is important. The 󰇝󰇞 where 9 is the highest possible value for .
Consistency reduces when  is lower or higher than  or equivalently 
, and most inequality happens when  and  have the maximum value (equal to ),
which results in . The 


, and given the highest inequality as a result of
assigning the maximum value by  and ,  is a value that should be subtracted from 
and  and added to , or equivalently shown in Eq. (24).
󰇛󰇜
(24)
As for the minimum consistency , Eq. (25) is given.
󰇛󰇜󰇛󰇜󰇛󰇜
󰇛󰇜
(25)

105
Solving for different values of 󰇝󰇞, the maximum possible  (max ) can
be found. The maximum values are used as CI, as indicated in Table 29. CI (max ) is found
by using Table 29, which lists the CI (max ) according to the  (Rezaei, 2015).
Table 29:  (max ) according to the  (Rezaei, 2015).

1
2
3
4
5
6
7
8
9
CI (max )
0
0.44
1.00
1.63
2.30
3.00
3.73
4.47
5.23
The CR is calculated using  and the corresponding CI, as shown in Eq. (26).


(26)
Since the consistency measurement proposed in the original BWM is based on which is the
optimal objective value (the output); it is called an Output-Based Consistency measurement.
CI is a consistency index, CR is a consistency ratio, and CR  [0, 1]. As much as CR is lower,
the comparisons are more consistent; therefore, the results are more reliable. Specifically, a CR
equal to zero means that the comparisons are cardinally consistent.
The solution space of Eq. (23) contains all positive values for ,  . The
weights sum is one, and the violation of all the weight ratios from their corresponding
comparison is a maximum of .
3.2.7.5.1.2 Input-based approach
According to Liang et al. (2020), unlike the Output-based Consistency Ratio, the Input-based
Consistency Ratio (CRI) can immediately show the level of consistency of decision-makers.
This is because instead of going through the whole optimization process, this approach uses
the input provided by the respondent, i.e., the respondent’s preferences. The equation relevant
to CRI is as follows.


(27)
Where
󰇱


(28)
 is the global input-based CR for all criteria,  indicates the local consistency
level associated with the criterion .  shows the preference for the best criterion over 
over criterion , j= 1, 2, …, n.  represents the preference for criterion  over the worst
criterion , j= 1, 2, …, n.  indicates the preference for the best criterion over the worst
criterion.
Input-based consistency measurement has advantages over output-based consistency
measurement. These advantages are mentioned according to Liang et al. (2020).

106
• Input-based consistency measurement can provide immediate feedback. Input-based
consistency measurement is based on input (preferences), meaning completing the
entire elicitation process is unnecessary. On the other hand, output-based consistency
measurement is based on output (weights), making it difficult to determine the level of
consistency. The simple input-based consistency measurement calculation makes it
easy to provide immediate feedback to the decision-makers.
• Its interpretation is simple: it is the maximum normalized discrepancy between the
value of  and its estimated value calculated as the indirect comparison .
• It can provide clear guidance to decision-makers on how to appeal inconsistent
judgment(s). The Output-based Consistency Ratio represents the level of global
consistency but cannot determine which judgments should be modified. However, the
Input-based Consistency Ratio demonstrates the levels of consistency related to
individual criteria. After determining the maximum local Input-based Consistency
Ratio, the most inconsistent judgment can be found, after which the decision-maker can
modify the judgments accordingly instead of revising without instructions.
• It is independent of the model. In other words, the Input-based Consistency Ratio can
be used independently to measure the consistency level in various BWM models, for
example, a non-linear or linear model or a multiplicative model (Brunelli and Rezaei,
2019). For instance, the linear BWM model does not have an effective consistency
measurement (Rezaei, 2016). Also, the non-linear BWM model (Rezaei, 2015) has a
different interpretation than the multiplicative BWM model (Brunelli and Rezaei,
2019). However, the CRI is the same in all three models. Therefore, input-based
consistency measurement does not depend on optimization models.
Considering the advantages of input-based consistency ratio () over the out-put
based consistency ratio,  is used in this study. Hence, it is important to know the
thresholds.
The algorithm for obtaining the threshold for the consistency ratio is shown below
(Liang et al., 2020).
1. Create pairwise comparison vectors (step 1). When there are  criteria 󰇛󰇜,
two random vectors 󰇛󰇜 and 󰇛󰇜 with the maximum
scale 󰇛󰇜, are generated to represent the vectors of the pairwise
comparison  and  in BWM. The elements in  and  are integers
randomly selected from the domain 󰇟󰇠.
2. Create the ordinal-consistent group (step 2). After generating a pair of vectors  and
, it is assigned to the ordinal-consistent group if it meets the ordinal consistency
condition, which is “(󰇜󰇛󰇜󰇛
󰇜”, and .
3. Create the ordinal-inconsistent group (step 3). If the paired vector created in Step 1 does
not meet the ordinal consistency condition, it is assigned to the ordinal-inconsistent
group and .

107
4. In step 4, continue to create the ordinal-consistent and ordinal-inconsistent groups
through steps 1–3 until the size of both groups is 10,000.
5. Step 5 calculates  for all paired vectors in these two groups using Eq. (27) and Eq.
(28).
6. In step 6, calculate the empirical cumulative distribution of  for the two groups
using Eq. (29) and Eq. (30).

󰇛󰇜
󰇥󰇦


(29)
Where 
󰇛󰇜 is the empirical cumulative distribution of  for the two groups. 󰇝󰇞 is
the indicator function shown in Eq. (30).  is the pair number of pairwise comparisons, 
is the  󰇛󰇝󰇞󰇜 input-based  obtained from this N pairs of preferences, 
󰇟󰇠 is the possible threshold.
󰇝󰇞

(30)
Where,
: the  󰇛󰇝󰇞󰇜 input-based 
: possible threshold
7. In step 7, calculate the relative rejected proportion of the s in the acceptable group
(
) and the accepted proportion of the s in the unacceptable group
(
) using Eq. (31) and Eq. (32).


󰇛󰇜

󰇛󰇜
󰇛󰇜
(31)


󰇛󰇜

󰇛󰇜
󰇛󰇜
(32)
Where,

: relative rejected proportion of the s in the acceptable group

: accepted proportion of the s in the unacceptable group
8. In step 8, if there is a  making 

 , then  is the threshold. If
not, go to the next step.
9. In step 9, specify the cross point of the lines of 
 and 
 , the  at this
point is used as the threshold.
The  thresholds according to the number of criteria and maximum value in the pairwise
comparison system () are listed in Table 30. The  values below the threshold are
acceptable.

108
Table 30:  thresholds based on the number of criteria and  (Liang et al., 2020).

Number of criteria
3
4
5
6
7
8
9
3
0.1667
0.1667
0.1667
0.1667
0.1667
0.1667
0.1667
4
0.1121
0.1529
0.1898
0.2206
0.2527
0.2577
0.2683
5
0.1354
0.1994
0.2306
0.2546
0.2716
0.2844
0.2960
6
0.1330
0.1990
0.2643
0.3044
0.3144
0.3221
0.3262
7
0.1294
0.2457
0.2819
0.3029
0.3144
0.3251
0.3403
8
0.1309
0.2521
0.2958
0.3154
0.3408
0.3620
0.3657
9
0.1359
0.2681
0.3062
0.3337
0.3517
0.3620
0.3662
3.2.7.5.2 BWM: post-optimality
If there are more than three criteria and a CR is greater than zero, Eq. (23) has multiple optimal
solutions. The upper and lower bounds of weights are acquired by solving Eq. (33) and Eq.
(34). Also, , which is on the right-hand side of the constraints of Eq. (23), is replaced by  in
Eq. (33) and Eq. (34) (Rezaei, 2016).

Subject to 






(33)

Subject to 






(34)
An individual chooses an optimal solution from the interval weights that could be, for example,
the center of the intervals.
3.2.7.5.3 Linear BWM, min-max
Eq. (23) could result in multiple optimal solutions. If, instead of minimizing the maximum
value among the set of {

}, minimizing of the maximum among the
set of 󰇝󰇞, the problem can be formulated as Eq. (35) (Rezaei,
2016).

󰇝󰇞
(35)

109
Subject to 


Eq. (35) is converted (converted min-max) to Eq. (36). As it is linear,  is denoted by 
(Rezaei, 2016).

Subject to 




(36)
Eq. (36) is an excellent linear approximation of Eq. (23). Hence, it offers a unique
solution to the problem (Rezaei, 2016). After solving the problem (10), the optimal weights
󰇛󰇜 and  are obtained.  can be considered directly as an indicator of the
consistency of the comparisons in this model. It should be noted that  , which is obtained
from Eq. (36), should not be divided by the values of the CI mentioned in Eq. (26). The closer
the value of  is to zero, the higher the consistency.
3.2.7.6 Finding the optimal weights: an alternative approach based on Bayesian BWM (A
group decision-making model)
BWM cannot integrate the preferences of multiple decision-makers into the so-called group
decision problem. Utilizing the average operator, for example, the geometric or arithmetic
mean, is a common way to aggregate the preferences of multiple decision-makers. Averages,
however, are sensitive to outliers and provide limited information about the overall preferences
of all decision-makers. Mohammadi and Rezaei (2020) developed a Bayesian hierarchical
model that can determine the optimal weights of a set of criteria according to the preferences
of multiple decision-makers utilizing the best-worst framework. BWM first gains the weight
of each decision-maker and then applies arithmetic mean to aggregate them. However, using
probabilistic modeling, Bayesian BWM calculates the aggregated distribution and all
individual preferences at once. The following is the description of Bayesian BWM based on
the article by Mohammadi and Rezaei (2020). The Bayesian BWM is a valid method to predict
the importance of criteria (Kalpoe, 2020a).
3.2.7.6.1 Group decision-making: a joint probability distribution
Assume that the  decision-maker, , evaluates the criteria  by providing
the vectors 
 and 
. The set of all vectors of K decision-makers is represented by 
 and

. The superscript 1: K demonstrates the total of all vectors in the base. In addition, the overall
optimal weight is denoted by .

110
Estimation  requires the use of several auxiliary variables. Specifically,  is
calculated according to the optimal weights of K decision-makers indicated by , 
. Therefore, the Bayesian model can compute  and  simultaneously. Before
making any statistical inference, it is required to write the joint probability distribution of all
random variables according to the available data. The 
 and 
 are given, and  and
 must be calculated accordingly in group decision-making within the BWM. Eq. (37)
indicates the joint probability distribution (Mohammadi and Rezaei, 2020).
󰇛

󰇜
(37)
Where,
󰇛

󰇜: joint probability distribution

 and 
: set of all vectors of K decision-makers
: overall optimal weight
: optimal weights of K decision-makers
After calculating the probability in Eq. (37), the probability of each variable can be estimated
utilizing Eq. (38) (Mohammadi and Rezaei, 2020).
󰇛󰇜󰇛󰇜

(38)
Where 󰇛󰇜 is the probability of each variable, and  and  are two arbitrary random variables.
3.2.7.6.2 Bayesian hierarchical model
In order to develop a Bayesian model, the independence and conditional independence of
variables need first to be recognized. Figure 10 illustrates the probabilistic graphical model of
the Bayesian BWM.

111
Figure 10: The probabilistic graphical model of the Bayesian BWM (Mohammadi and
Rezaei, 2020).
In figure 10, the nodes are the variables. Also, rectangles are the observed variables that are
the original BWM inputs. Besides, circular nodes are variables that require to be calculated.
Further, arrows indicate that the node at the origin depends on the node at the other end. This
means that the value of  depends on 
 and 
, and the value of  depends on .
The plate that covers a set of variables implies that the corresponding variables are
iterated for each decision-maker. There is no  on the plate because there is only one 
for all decision-makers.
The conditional independence between various variables is clear based on Fig. 6. For
example, 
 is independent of  given  i.e., Eq. (39) (Mohammadi and Rezaei, 2020).



(39)
Where,
󰇛

󰇜: joint probability distribution

 and 
: set of all vectors of K decision-makers
: overall optimal weight
: optimal weights of K decision-makers

112
Taking into account all the independence between the various variables, the application
of the Bayes rule for the joint probability (Eq. (37)) leads to Eq. (40) (Mohammadi and Rezaei,
2020).
󰇛

󰇜󰇛

󰇜󰇛󰇜
󰇛󰇜

󰇛󰇜


(40)
where the last equality is achieved utilizing the probability chain rule and conditional
independence of various variables, and each decision-maker independently presents the
preferences. There is a chain between different parameters because the calculation of the
parameters in Eq. (40) relies on other variables. The chain is the reason for being called a
hierarchical model.  and  can be well modeled utilizing the multinomial distribution,
meaning that they retain the original idea of BWM. It is important to note that  indicates the
preference of all criteria over the worst criterion, while  shows the preference of the best
criterion over other criteria. Therefore, they can be modeled as shown in Eq. (41) (Mohammadi
and Rezaei, 2020).

󰇻󰇧
󰇨

󰇻
(41)
The multinomial represents a multinomial distribution. Given  one can expect
each  to be in its vicinity. For this purpose, the Dirichlet distribution is re-parametrized
concerning its mean and concentration parameter. Eq. (42) presents the models  given 
(Mohammadi and Rezaei, 2020).
󰇛󰇜
(42)
Where
: mean of the distribution
: concentration parameter
Eq. (42) stated that the weight vector  associated with each decision-maker must be
adjacent to  because it is the mean of the distribution, and the non-negative parameter γ
controls their proximity. This technique is applied to various Bayesian models (Kruschke,
2014). The concentration parameter should also be modeled utilizing the distribution. Eq. (43)
gives a reliable option: the gamma distribution satisfies the non-negativity constraints
(Mohammadi and Rezaei, 2020).
󰇛󰇜
(43)
Where

113
 and : Gamma distribution shape parameters.
The previous distribution on  is shown utilizing an uninformative Dirichlet
distribution with the parameter  in Eq. (44) (Mohammadi and Rezaei, 2020).
󰇛󰇜
(44)
As the determined model does not bear a closed-form solution, the Markov-chain
Monte Carlo (MCMC) technique is utilized to calculate the posterior distribution. For the
MCMC sampling, “Just Another Gibbs Sampler” (JAGS) is utilized (Plummer, 2004), which
is a probabilistic language for sampling and posterior computation (Forman and Peniwati,
1998). Hence, the model's output is the posterior distribution of weights for each decision-
maker and the aggregated  (Mohammadi and Rezaei, 2020).
3.2.7.6.3 Credal ranking
The Bayesian BWM brings forward the credal ranking concept to measure the relationship
between a pair of main-criteria or sub-criteria (Mohammadi and Rezaei, 2020). Compared to
the traditional method, which utilizes only two figures to specify the superiority of confidence,
it can design a Bayesian test in order to calculate the confidence of each credal ranking. By
employing this principle in the real-world case, the superiority of confidence between different
pairs of competence criteria can be calculated (Li et al., 2020). Credal ranking can calibrate the
degree of superiority of one criterion over another. The posterior distribution of weights assists
in measuring the confidence of the relationships between different criteria. A weighted directed
graph visualizes the credal ranking based on which the interrelation of criteria and confidences
are merely understood. In this graph, each node represents a criterion, and each edge indicates
the obtained confidence. Eq. (45) describes the credal ordering  , for a pair of criteria  and
 (Mohammadi and Rezaei, 2020).
󰇛󰇜
(45)
Where
: the relationship between the criteria  and , i.e., ,  , or  ;
󰇟󰇠 : confidences of the relationship
For a set of criteria 󰇛󰇜 , the credal ranking is a set of credal orderings that
contains all pairs 󰇛󰇜, for all .
Confidence in the credal ordering can offer more information to decision-makers who
can make better decisions in particular. Eq. (46) provides a Bayesian test according to which
the confidence of each credal ordering can be calculated (Mohammadi and Rezaei, 2020).
󰇛󰇜󰇛󰇜
(46)

114
Where
󰇛󰇜: posterior distribution of 
󰇥

This integration can be estimated from the Markov-chain Monte Carlo (MCMC)
samples. Having  samples of the posterior distribution, the confidence can be calculated as
shown in Eq. (47) (Mohammadi and Rezaei, 2020).

󰇛󰇜



󰇛󰇜


(47)
Where
:  sample of  from the MCMC samples.
Therefore, one can calculate the confidence of superiority (confidence level) over the
other for each pair of criteria. Credal ranking can be changed to traditional one (the common
way of ranking criteria): since ,  is more important than  if
and only if . The traditional ranking of criteria can be achieved by setting a
threshold of 0.5 for credal ranking. The closer the Confidence Level (CL) is to 1, the more
pronounced the degree of certainty about the relation, which indicates that one criterion is
certainly considered more important than another (Mohammadi and Rezaei, 2020).
It is important to note that the credal ranking can be changed into the conventional
ranking merely by applying the threshold of 0.5 to the obtained confidence. However, the
threshold can vary from problem to problem, and choosing a particular threshold value is
entirely up to the decision-maker. In other words, credal ranking can be shaped so that they
show the ranking of criteria in various problems based on the confidence desired by decision-
makers (Mohammadi and Rezaei, 2020).
3.2.7.6.4 Introducing the CL classification in the credal ranking (Bayesian BWM)
There is no specific classification to describe CL in the literature. Hence, this study intends to
introduce the CL classification to explain the results according to the previous studies (Kalpoe,
2020; Li et al., 2020; Mohammadi and Rezaei, 2020). In this regard, Table 31 introduces a
description of each CL range for a threshold value of 0.5.
Table 31: Description for each CL range for a threshold value of 50.
CL range
Description
0.8 ≤ CL
One criterion is certainly more important than the other
0.60 ≤ CL < 0.80
One criterion is more important than another
0.50 ≤ CL < 0.60
Superiority of one criterion over another is not well established

115
It should be noted that when the threshold value is 0.5, values less than 0.5 are not
considered in this classification because values less than 0.5 () must be interpreted
inversely. For instance, when the confidence level for comparing C1 and C2 is 0.30, C2 is more
important than C1, with a confidence of 0.7 (i.e., ).c
3.3 Comparative analysis and selection of the MCDM method that
will be used
MCDM methods can be compared with the following criteria according to the literature
(Brispat, 2017).
• Year of development or method proposal: This aspect is only intended to clarify the age
of the method. The advantage of using the method developed a long time ago is that it
has been used for a long time, offering reliability and results. Nevertheless, time
changes and a younger method may be more useful in this dynamic environment than
the old methods.
• Transparency of the method: Extent and ease of understanding of the method. Some
methods are challenging to understand, while others are easy. This criterion indicates
whether this method is easily understood and hence easily applicable.
• Required data: The amount of data required is also an important factor to consider. The
fewer data needed to achieve reliable results, the more points the method scores in this
area.
• Quality of the weights: This is used to evaluate the result of pairwise comparison.
• Ability to combine with other methods: The ability to combine with other MCDM
methods
• Avoid equalizing bias: Equalizing bias refers to a condition in which the individual
gives (approximately) the same weight to all the decision-making attributes (Fox and
Clemen, 2005; Tervonen et al., 2017, Marttunen et al., 2018).
The two main categories of information required by PROMETHEE are the weight of the
criteria and the preference of decision-makers if any. In other words, there is no particular
method for determining weight, which can be considered a disadvantage. In addition, dealing
with more criteria (eight or higher) can make the situation difficult for the decision-maker
(Serrai et al., 2017). This makes it challenging to achieve a reliable and realistic perception of
the stakeholders. Finally, transparency can be classified at a very low level due to difficulty.
Also, the transparency of the ELECTRE method is very low due to the comprehensive
description.
TOPSIS method is complex and takes time to understand, resulting in low transparency.
Because using the Euclidean distance, any correlation between the criteria is not considered,
and the qualitative weight parameters may be problematic (Sarai et al., 2017).
In general, understanding and implementing the AHP method is not much complex. With
the four steps (Saaty, 1994; Bian et al., 2017), the transparency of the AHP method is at the
same level as the best-worst method. Therefore, the transparency of BWM can be considered

116
in the middle category and not in the high transparency category. It should also be noted that
study results by (Rezaei et al., 2021) pointed out that AHP and BWM have a low equalizing
bias.
On the other hand, WSM and WPM methods are easy to understand and do, which leads to
very high transparency.
To evaluate MCDM methods, Table 32 summarizes their benefits, drawbacks, and features
according to the literature (Brispat, 2017, Rezaei et al., 2021) and our analysis.
Table 32: Evaluation of MCDM methods.
MCDM
Year of development
or method proposal
Transparency of
the method
Required
data
Quality of
weights
Ability to
combine with
other methods
Avoid
equalizing
bias
ELECTRE
1966
Very negative
Neutral
Positive
Positive
Not available
WSM
1967
Very positive
Very
positive
Not available
or very
negative
Very positive
Not available
WPM
1969
Very positive
Very
positive
Not available
or very
negative
Very positive
Not available
AHP
1980
Neutral
Neutral
Positive
Neutral
Positive
TOPSIS
1981
Negative
Neutral
Not available
or very
negative
Not available or
very negative
Not available
PROMETHEE
1985
Very negative
Neutral
Positive
Negative
Not available
BWM
2015
Neutral
Very
positive
Very positive
Very positive
Positive
The MAMCA method explicitly considers the interests of different stakeholders in the
analysis. Therefore, one of the essential parts of the decision-making process is paying attention
to the stakeholders’ interests. For this reason, MAMCA has been chosen as a way to determine
perception. Especially, MAMCA is a method that is not difficult to understand and has seven
steps; therefore, the transparency of this method can be placed in the middle category. Step 3
MAMCA analysis is to determine the important criteria and their weight. Then, using another
MCDM method combined with the MAMCA analysis, weights can be assigned to criteria.
Weight allocation requires a comparison method that allows a fair and accurate comparison of
criteria. More accurate results can be obtained using pairwise comparisons because only two
factors are compared at a time.
Nevertheless, most pairwise comparison methods, such as ELECTRE, PROMOTHEE,
and AHP, cannot resolve recurring inconsistencies. The BWM method uses a different pairwise
comparison and makes more consistent results possible with less information. The weights are
determined by comparing the best criterion with the rest and other criteria against the worst
criterion. It is essential to notice that AHP requires the pairwise comparison of all n decision
criteria, i.e., 󰇛󰇜
 pairwise comparisons. On the contrary, BWM requires only the so-called
reference pairwise comparisons, i.e.,  pairwise comparisons (Liang et al., 2020). In
addition, the special structure of BWM generates two vectors comprising only integers, which
avoids a fundamental distance problem related to the fractions used in pairwise comparisons
(Salo and Hämäläinen, 1997).

117
BWM seems to be one of the best ways to decide on the weight of parameters (Serrai
et al., 2017). This is because the users predefined the best and worst criteria and the
comparisons of other elements. In addition, this method is not difficult to understand (average),
and the need for fewer data makes this method attractive to use. Besides, BWM has a low
equalizing bias. BWM is an easy-to-apply and easy-to-understand approach that makes the
comparisons structured and results in more consistent comparisons, thus, more reliable
weights/rankings. It is appropriate for both conditions when flexibility is not desirable (linear
BWM), and flexibility is desirable (nonlinear BWM). Suitable for both group and individual
decision-making. Supports reaching consensus in a natural way. It is efficient in terms of input
data. It can be used for various MCDM problems with quantitative and qualitative criteria.
Finally, it is compatible with many other MCDM approaches.
To conclude and summarize the above analysis, the appropriate method for conducting
the analysis, considering the various stakeholders, is MAMCA. The third step of the MAMCA
is to determine the main criteria and weights. This means that another method is required in
order to determine the essential criteria and weights for comparing alternatives. This chapter
analyzed popular MCDM methods and reported which method will perform PBA. BWM
(Bayesian BWM) is the only method with a very high quality of weight (described in section
3.2.7.6 of Chapter 3) and requires a small amount of data. Also, it has a low equalizing bias.
Also, the other advantages of this method include the combination of weight quality, fewer
inconsistencies between criteria, fewer data required to obtain highly reliable results, and
average transparency of the method. Bayesian BWM is used in this study because different
groups of stakeholders are involved. Before calculating the optimal group weights by Bayesian
BWM, the consistency of the respondents can be examined using the Input-based approach
(Eq. (27 and 28) in section 3.2.7.5.1.2 of Chapter 3), and acceptable ones (their obtained global
input-based consistency ratio is less than the input-based consistency ratio thresholds) can be
considered (Liang et al., 2020). After eliminating pairwise comparisons with unacceptable
consistency ratios (section 3.2.7.5.1.2 of Chapter 3), different sample sizes can be obtained and
utilized for different levels of the model. Also, it is important to note that Bayesian BWM can
provide much more information than the original BWM. Bayesian BWM can provide the credal
ranking and confidence level in the weight-directed graph. This helps to understand the
importance perceived by stakeholders of one criterion over other criteria.

118
Chapter 4
Method Implementation
After choosing MAMCA and Bayesian BWM as the methods used for this study (section 3.3
of Chapter 3), this section applies such methodologies for determining each stakeholder's
perception of the importance of criteria. However, it is important to note that after determining
unacceptable data using the input-based approach (described in 3.2.7.5.1.2 of Chapter 3) and
excluding them from the analysis (section 5.4.3 of Chapter 5), the rest data can be used in
Bayesian BWM (explained in section 3.2.7.6 of Chapter 3) to find the importance of each
criterion.
The first step in MAMCA is defining the problem and identifying the alternatives
covered in Section 4.1. Step 2 of MAMCA is the stakeholder analysis explained in section 4.2,
which aims to describe the important stakeholders in shared mobility systems. Then the
selection of criteria is described in section 4.3 as follows.
4.1. Problem definition and alternatives selection
The first step of the MAMCA (mentioned in section 3.1.1 of Chapter 3) implies defining the
problem and classifying the possible alternatives. The problem is identifying the gap between
different stakeholders' needs, expectations, and perspectives of the important shared mobility
services. In this regard, the research questions are mentioned in Chapter 1. This study focuses
on the main modes of shared mobility that are available in Turin (the selection of Turin as the
case study is mentioned in section 5.1.2 of Chapter 5) at the time of writing: car-sharing, bike-
sharing, and scooter-sharing (Sharing di Monopattini Elettrici) (Comune Torino, 2021).
Information about shared mobility services in Turin is mentioned in section 5.1.2 of Chapter
5.

119
4.2 Stakeholder analysis
The second step of the MAMCA (mentioned in section 3.1.2 of Chapter 3) is to perform a
stakeholder analysis to recognize the stakeholders involved. Macharis et al. (2010) define
stakeholders as those interested in or influenced by decisions made during the process.
Stakeholder analysis can be performed to visualize stakeholder alignment and demonstrate
common and conflicting interests. This helps to minimize project threats and barriers and
maximize collaboration.
The shared mobility system user group is an important stakeholder to be considered (Jia
et al., 2018) and directly affects the demand for these services. Operators are also major
shareholders. Operators invest in vehicles and infrastructure, system operations, and day-to-
day operations management (Zhang et al., 2105). The public authorities are another important
stakeholder (Lan et al., 2017). Public authorities are called government members in this study,
including three levels. The first level contains regional executive directors and staff. The
second level includes metropolitan city executive directors and staff. Finally, the third level
comprises municipal policy-makers, executive directors, and staff. The government can control
the norms, policies, and regulations, such as limiting the number of shared mobility vehicles
and dividing parking lots by local authorities. Another example is providing regulations for
developing the shared mobility industry at the national level (Miller et al., 2016). In this regard,
Zhang et al. (2105) showed that government participation and bike-sharing companies’
investment in operations management are of considerable importance in the sustainable
development of the bike-sharing industry. Also, the government could improve the legal
framework for creating dedicated parking spaces for car-sharing vehicles on public streets by
redesigning road traffic regulations. Also, municipalities can improve pedestrian, bicycle, and
public transportation infrastructure as complementary modes of transportation for car-sharing.
In addition, they can install reserved parking spaces for shared vehicles in crowded cities or
near public transportation junctions and limit motorized traffic within cities (Loose et al.,
2006).
Furthermore, the government can also improve media communication efforts to
influence user behavior (Jia et al., 2018). On the other hand, a shared mobility system can
benefit the government and the people. For instance, by e-scooter-sharing development,
governments can lead to developing sustainably by addressing development problems such as
pollution and traffic during rush hour (Ling et al., 2015; Axsen and Sovacool, 2019). Hence,
the most relevant stakeholders to the shared mobility systems are operators, government
members, and passengers (Turoń et al., 2020). Having a better understanding of the views of
these stakeholders and extensive interactions may improve the state of the shared mobility
system. In addition, the development of sustainable urban mobility plans or new sustainable
transport regulations may influence local transport policymakers (Dörry and Decoville, 2016;
Le Pira et al., 2016; Le Pira et al., 2017). This study also considers the non-users of shared
mobility systems in order to understand their perception. This can help to understand the gap
(if any) in the views of users and non-users of shared mobility services, which can help to
provide some policies to attract them to use these services and increase demand. For a better

120
view of the stakeholders involved in this study, the important stakeholders of shared mobility
services and their relationship are shown in Figures 11 and 12, respectively.

121
Figure 11: Important stakeholders of shared mobility services.
Figure 12: Relationship between the stakeholders of shared mobility services.

122
4.3 Selection of criteria
This study considers four shared mobility service stakeholders: users, non-users, government
members, and operators. In this section, the selection of criteria is described. The third step of
the MAMCA (mentioned in section 3.1.3 of Chapter 3) is the selection of important criteria,
and weights for stakeholders, including users/non-users (common criteria for users and non-
users are used to find the gap in their opinions about the importance of the criteria), government
members and operators explained in sections 5.3.1, 5.3.2 and 5.3.3, respectively. It is important
to note that as mentioned in section 3.1.3 of Chapter 3, in order to show the stakeholders
involved with their goals and objectives, it is possible to provide a hierarchical criteria tree (at
this stage); however, it is not used in this study since it is not among the research purpose. In
this regard, it can be noted that the criteria selection is based on the objectives of the
stakeholders involved and according to the considered alternatives (car-sharing, bike-sharing,
and scooter-sharing). Also, as stated in section 3.1.3 of Chapter 3, it is possible to assign
weights to stakeholders if necessary. These can show the importance of stakeholders in the
decision-making process contributing to determining the importance of the criteria from all
stakeholders' views simultaneously (the overall importance of each criterion according to the
combinations of perspectives). However, it is not used in this study because, in this section, the
purpose of the study is to determine the point of view of each stakeholder separately. According
to the research objectives, this research has two parts, including an analysis of shared mobility
services (as a whole, not for a specific shared mobility service) and an analysis of each shared
mobility service (separately), as follows.
• Analysis of shared mobility services (as a whole, not for a specific shared mobility
service): in this part, the perspectives of stakeholders (users, non-users, government
members, and operators) of shared mobility services (as a whole, not for a specific shared
mobility service) about the importance of the criteria associated with each stakeholder is
determined. In other words, users and non-users determine the importance of each criterion
(associated with their perspectives) depending on the extent to which it motivates them to
use (more use) shared services. Members of the government specify the importance of each
criterion (relevant to their perspective) when a new shared mobility system is launched in
Turin, Italy. Also, Operators determine the importance of each criterion (related to their
perspective) to the extent that it can motivate them to implement their shared mobility
system in Turin. The importance of each criterion (weight) can be found using Bayesian
BWM (explained in section 3.2.7.6 of Chapter 3). Besides, at the end of this part, since data
on users' and non-users' opinions on the value of each criterion (indicator value, explained
in section 3.1.4 of Chapter 3) are also collected (presented in Chapter 5), the preferred shared
mobility service (car-sharing, bike-sharing, and scooter-sharing) from the perspectives of
users and non-users groups can be determined (using step five through seven of MAMCA,
mentioned in 3.1.1.5, 3.1.1.6, and 3.1.1.7 of Chapter 3). Also, the gap between (if any)
perceptions of users and non-users (perception analysis) can be found. Further, sensitivity
analysis and scenarios can be done from users’ and non-users' perspectives.
• Analysis of each shared mobility service (separately): this part determines the perspective
of each stakeholder (users, non-users, government members, and operators) of car-sharing,

123
bike-sharing, and scooter-sharing services on the importance of each criterion and sub-
criterion affecting passengers' shared mobility choice behavior. These criteria and sub-
criteria are the same among the stakeholders of all three shared mobility services. This helps
to find differences in their views (if any) about the importance of each criterion and sub-
criterion. The importance of each criterion (weight) can be found using Bayesian BWM.
In order to have a better understanding of the study purpose of these two parts, Figure 13 shows
the purpose of each part separately.
Figure 13: Purpose of analysis of shared mobility services (as a whole, not for a specific
shared mobility service) and an analysis of each shared mobility service (separately).
4.3.1. Analysis of perspectives of stakeholders of shared mobility services (as
a whole, not for a specific shared mobility service)
According to the description of the analysis of shared mobility services (as a whole, not for a
specific shared mobility service), this analysis can be divided into three sub-sections. The sub-

124
section 4.3.1.1 identifies the influential characteristics in choosing a shared mobility service
for a trip for users and non-users. Also, in sub-section 4.3.1.2, important characteristics for
government members when a new shared mobility system is set up in Turin, Italy, are specified.
In addition, sub-section 4.3.1.3 sets out characteristics that can be considered important
elements for the shared mobility operators to implement the shared mobility system in a city.
4.3.1.1 User and non-user perspectives about shared mobility services (as a whole, not for
a specific shared mobility service) perspectives
The main characteristics that users and non-users of shared mobility services (as a whole, not
for a specific shared mobility service) consider when selecting a shared mobility service to
make a trip are listed below in a similar study (Brispat, 2017) and also author’s knowledge
summarized in section 2.4. It is important to note that in this study, the 7-point Likert scale is
used to measure the respondent’s opinions about the criteria. According to Khandelwal (2021),
this scale is the most accurate Likert scale because it best represents the respondent’s feelings.
Therefore, it provides better accuracy in results and is very useful for researchers. However, it
should be noted that the 7-point Likert items suffer from bias in response style. The definition
and explanation of the measurement of the criteria used in this research are as follows. Again
according to Brispat (2017) and knowledge of the author, items in the list are sorted from the
most important to the least important:
• Accessibility: ease of access, availability of a shared vehicle, proximity to the location
of the parked shared vehicle. For travelers, this aspect may occupy an important place
in selecting a shared mobility system. It is essential to figure out how easy or difficult
it is for passengers to access these shared mobility services. Passenger safety can be
defined between [1-7], which means 1 is very difficult, and 7 is very easy.
• Cost: expenses for shared mobility usage, such as service subscription fees or usage
fees. Ticket prices can be the main aspect to consider. For instance, travelers are more
likely to opt for cheaper shared mobility services. Therefore, it is essential to determine
how passengers rate the cost of usage or membership fee. It can be measured in degrees
between [1-7], meaning that 1 is very expensive, and 7 is very cheap.
• Comfort: vehicle characteristics that make passengers feel comfortable during the trip.
It can vary between shared transportation services; hence, travelers may prefer a shared
transportation service based on travel comfort. Hence, it is required to know how
comfortable passengers feel on each trip of the shared transport service. It can be
measured in degrees between [1-7], meaning 1 is very uncomfortable, and 7 is very
uncomfortable.
• Travel Safety: the level of safety of the individuals during the trip, such as the rate of
accidents, harassment, assault, and theft. Passenger safety information provides insight
into how safe a passenger feels when using a shared transportation service. These safety
measures can be different within the service and have a different sense of safety. For
instance, travelers who use a shared transportation service perceive safety as a
perception or feeling of safety. Therefore, it is important to determine how safe the
passenger feels with each shared mobility service. In this case, passenger safety can be
defined between [1-7], which means 1 is very unsafe, and 7 is very safe.

125
• Operational speed: the average velocity that a shared mobility system overpasses. It
should be specified how passengers would rate the travel speed of each shared mobility
service. It can be measured in degrees between [1-7], which means 1 is very poor and
7 is very good.
• User-friendliness: easy for beginners to learn, easy to use, and provide travel
information in the app. To understand how easy or difficult it is for passengers to access
any shared mobility service, this characteristic can be defined between [1-7], which
means that 1 is very difficult, and 7 is very easy.
• Image: the image of a shared mobility system in passengers' eyes. It is expected, for
example, that the image of a car-sharing service differs from a bike-sharing service or
a scooter-sharing service. Hence, it is important to know how passengers would rate
each shared mobility service overall. The image of a system can be measured in degrees
between [1-7], which means that 1 is very poor and 7 is very good.
• Possibility of carrying items: possibility of carrying luggage or bags or shopping items
in the shared vehicle. For instance, passengers can carry their luggage by shared car but
not by scooter-sharing. Thus, it is necessary to know whether it is difficult or easy for
passengers to carry belongings when using any shared mobility service. It can be
defined between [1-7], which means 1 is very difficult, and 7 is very easy.
The characteristics studied differ for government members, operators, and people
(users/non-users) of shared mobility services (as a whole, not for a specific shared mobility
service). Therefore, it is better to indicate them with different symbols. In this regard, “Cp”
denotes the criteria related to both users and non-users. These symbols are presented in Table
33.
Table 33: Symbolize each criterion associated with users and non-users.
Criteria
Symbols
People safety
Cp1
Operational speed
Cp2
Accessibility
Cp3
User-friendliness
Cp4
Image
Cp5
Comfort
Cp6
Cost
Cp7
Possibility of carrying items
Cp8
4.3.1.2 Government perspective about shared mobility services (as a whole, not for a
specific shared mobility service)
It is remarkable to know the views of government members on some of the features associated
with shared mobility services (as a whole, not for a specific shared mobility service) that may
be important to members of the government. These criteria are presented according to a similar
study (Brispat, 2017), and the knowledge of the author is listed from the most important to the
least important.
• Average number of trips per vehicle per day: it gives insight into the efficiency of
the vehicle that shows the efficiency of the service.
• Greenhouse gases (GHGs): the amount of greenhouse gas emissions by a shared
mobility system.

126
• Parking issues: illegal parking of shared vehicles like parking in inappropriate places.
• Emission of pollutants (CO2/km): pollutants emitted by a shared vehicle.
Governmental members care about sustainability and strive for fewer emissions.
• Integration of the shared mobility service with public transport: complementarity
of a shared vehicle for public transport. Their integration can increase urban mobility.
• Vehicle fee (Euro): the fee that a shared mobility operator may pay to the municipality.
For example, car-sharing operators paid a fee to the municipality, which allowed their
shared cars to go to city centers or places where traffic was restricted.
In terms of the notation of characteristics, “Cg” denotes criteria related to the
stakeholder group of government members. These symbols are presented in Table 34.
Table 34: Symbolize each criterion associated with government members.
Criteria
Symbols
Average number of trips per vehicle per day
Cg1
Greenhouse gases (GHGs)
Cg2
Parking issues
Cg3
Emission of pollutants
Cg4
Integration of the shared mobility service with public transport
Cg5
Vehicle fee
Cg6
4.3.1.3 Operators' perspectives about shared mobility services (as a whole, not for a
specific shared mobility service) perspectives
The following characteristics can be considered important elements for system implementation
for a shared mobility operator planning to run the shared mobility system in a city. These
criteria are presented according to a similar study (Brispat (2017)), and the knowledge of the
author is listed from the most important to the least important.
• Vehicle utilization rate (%): total time (minutes) that all shared vehicles are used each
day divided by the time they can potentially be used per day in 24 hours, which shows
the efficiency of the service.
• Usage fees (membership fees) (€): operators experience higher revenue with higher
usage fees (membership fees), and it affects earnings.
• Average number of trips per vehicle per day: it gives insight into the efficiency of
the vehicle that shows the efficiency of the service.
• Operational speed (Km/h): the average velocity a shared mobility system passes.
• The lifespan of the vehicle (year): system lifespan is measured in years and is
indicated by the lifespan of vehicles.
In terms of the notation of characteristics, “Co” represents criteria related to the
stakeholder group of operators. These symbols are offered in Table 35.
Table 35: Symbolize each criterion associated with operators.
Criteria
Symbols
Vehicle utilization rate
Co1
Usage fees
Co2
Average number of trips per vehicle per day
Co3
Operational speed
Co4
The life span of the vehicle
Co5

127
4.3.2. Criteria related to traveler choices that are common across
stakeholders and shared mobility services
The characteristics that different stakeholders (government members, operators, and users and
non-users) rank by importance for each shared mobility service (car-sharing, bike-sharing, and
scooter-sharing) are all related to the passenger choices on whether to use the service or not,
and they should be the same. According to the literature (Chapter 2) and the author’s
knowledge, twelve important characteristics can affect shared mobility services. These
characteristics include travel time, travel distance, departure time, trip purpose, cost, comfort,
safety, service quality, environment-friendly system, user-friendliness, service availability,
vehicle availability and accessibility. As explained in section 3.2.7.1 of Chapter 3, when the
number of criteria is more than nine, if possible, they can be classified into different groups
since, generally, humans can only compare seven ± two attributes (Miller and Starr, 1963;
Glassman et al., 1994). According to the literature (Chapter 2), these important characteristics
can be divided into trip-related characteristics, service-related characteristics, and availability
and accessibility as follows (listed from the most important to the least important.).
4.3.2.1 Trip-related characteristics
Individuals could consider some trip-related characteristics in selecting each shared mobility
service to make a trip. These characteristics are listed below (listed from the most important to
the least important).
• Travel time: the time it takes with a given means to travel from origin to destination.
Stakeholders should be asked to specify which characteristics, including short-time
trips (less than 30 min), long-distance trips (beyond 30 min), or both, might drive people
to use (or use more) each shared mobility service.
• Travel distance: the distance between origin and destination. It is important to ask
stakeholders to identify which characteristics, including short-distance travel (less than
5 km), long-distance travel (beyond 5 km), or both, might induce people to use (or use
more) each shared mobility service.
• Departure time: the trip's start time, such as in the morning or evening, on weekends,
or on weekdays, during peak or off-peak hours. It is required to ask stakeholders to
specify which characteristics, including peak hours, off-peak hours, or both, might
encourage individuals to use (or use more) each shared mobility service. Also, it is
required to ask stakeholders to specify which characteristics, including traveling on a
weekday morning, on a weekend morning, on a weekday evening, or/and a weekend
evening might induce people to use (or use more) each shared mobility service.
• Trip purpose: the purpose of the trip, such as traveling to work, school, shopping, or
meeting a friend. Stakeholders should be asked to determine which characteristics,
including travel for leisure trips (e.g., visiting friends or shopping), non-leisure trips
(going to work/school), or both, might induce people to use (or use more) each shared
mobility service.

128
4.3.2.2. Service-related characteristics
Some characteristics of each shared mobility service affect people's behavior in choosing each
shared mobility service for travel. These characteristics are listed below (listed from the most
important to the least important).
• Cost: expenses for each shared mobility service usage, such as service subscription fees
or usage fees.
• Comfort: vehicle characteristics that make you feel comfortable during the trip.
• Safety: the level of safety of the individual during the trip, such as the rate of accidents,
harassment, assault, and theft.
• Service quality: quality of each shared mobility system and given services.
• Environment-friendly system: a system that reduces environmental impacts.
• User-friendliness: easy for beginners to learn, easy to use, and provide travel
information in the app.
4.3.2.3 Availability and accessibility
The definitions of two characteristics, including the availability and accessibility of each shared
mobility service that influence each shared mobility service demand, are as follows (listed from
the most important to the least important).
• Service availability: availability of each shared mobility service around shopping
malls, colleges, transportation centers, city centers, and densely populated areas.
• Vehicle availability and accessibility: availability of the vehicle where I need it,
easiness to reach and access the vehicle, proximity to the location of the parked vehicle
from my starting point.
4.3.3. Summary of the main-criteria and sub-criteria to be considered
In summary, Table 36 presents the three main-criteria and twelve sub-criteria that are
common across stakeholders and shared mobility services in analyzing each shared mobility
service (separately).
Table 36: The three main-criteria and twelve sub-criteria that are common across
stakeholders and shared mobility services.
Main-criteria
Sub-criteria
C1. Trip-related characteristics
C1.1. Travel time
C1.2. Travel distance
C1.3. Departure time
C1.4. Trip purpose
C2. Car-sharing characteristics
C2.1. Travel cost
C2.2. Travel comfort
C2.3. Safety
C2.4. Service quality
C2.5. Environment-friendly system
C2.6. User-friendly
C3. Availability and accessibility
C3.1. Service availability
C3.2. Vehicle availability and accessibility

129
Chapter 5
Experimental Activities
This section is dedicated to the experimental activities of this study. After the problem
definition and alternatives selection (step 1 of MAMCA, mentioned in section 3.1.1 of Chapter
3 and 4.1 of chapter 4), stakeholder analysis (step 2 of MAMCA, given in section 3.1.2 of
Chapter 3 and section 4.2 of Chapter 4), selection of the criteria for each study purpose (step 3
of MAMCA, presented in section 3.1.3 of Chapter 3 and section 4.3 of chapter 4), to obtain the
weight of each criterion (explained in section 4.3. of Chapter 4), first, the required data should
be gathered. To do this, the study area must be well explained, given in section 5.1. Then, all
the information related to questionnaire design, data collection activities, and collected data are
offered in sections 5.2, 5.3, and 5.4, respectively.
5.1 Study area
5.1.1. Shared mobility services in Italy
To better understand the shared transport services in Italy, it is better to explain their evolution
in chronological order, based on Ciuffini et al. (2021), a national report on shared mobility
released almost every year.
According to Ciuffini et al. (2021), the first Station-based bike-sharing was born in
Ravenna in 2000, followed by the first Station-based car-sharing in Milan in 2001. Later came
Free-floating car-sharing services that overlapped the previous station-based services, followed
by the new Free-floating bike-sharing and scooter-sharing services. The first Free-floating car-
sharing service was launched in Milan in 2013. In 2015, electric Free-floating car-sharing was
launched. Also, the first Free-floating bike-sharing service was introduced in Italy in 2016.
Besides, the Free-floating scooter-sharing service was launched in Italy in late 2019 and early
2020. Demand for car-sharing, bike-sharing, and scooter-sharing in 2020 is 6.4 million rentals
(down 48% from 2019 due to Covid-19), 5.7 million rentals (down 55% from 2019 due to
Covid-19), and 7.4 million rentals, respectively. The number and percentage of available

130
services, available vehicles, and rentals of each shared mobility system in Italy are illustrated
in Figures 14 to 16, respectively (Ciuffini et al., 2021).
Figure 14: Number and percentage of available services of each shared mobility system in
Italy in 2020 (Ciuffini et al., 2021).
Figure 15: Number and percentage of available vehicles of each shared mobility system in
Italy in 2020 (Ciuffini et al., 2021).
Free-floating Car-
sharing, 22 (15%)
Station-based
Car-sharing, 19
(13%)
Free-floating
bike-sharing, 14
(10%)
Station-based
bike-sharing, 25
(17%)
Scooter-sharing,
64 (45%)
The number and percentage of available services
Free-floating Car-sharing Station-based Car-sharing
Free-floating bike-sharing Station-based bike-sharing
Scooter-sharing
Free-floating Car-
sharing, 1293
(3%)
Station-based
Car-sharing, 5989
(13%)
Free-floating
bike-sharing,
24764 (54%)
Station-based
bike-sharing,
9941 (22%)
Scooter-sharing,
3555 (8%)
The number and percentage of available vehicles
Free-floating Car-sharing Station-based Car-sharing
Free-floating bike-sharing Station-based bike-sharing
Scooter-sharing

131
Figure 16: Number and percentage of rentals of each shared mobility system in Italy in 2020
(Ciuffini et al., 2021).
Furthermore, in order to better understand the diffusion of shared mobility services in Italy,
knowing the number of subscribers of each shared transportation service where this service is
available in Italy (Ciuffini et al., 2021) and the population
3
of the province and city
4
, the ratio
of the subscribers of each shared mobility service to the population of the province and city (in
percentage) is obtained, which is presented in Table 37.
Table 37: The ratio of the subscribers of each shared mobility service to the population of the
province and city (Ciuffini et al., 2021).
Type of shared
mobility service
City
Name
Province
population
City
population
Number of service
subscribers in 2020
Subscribers/province
population ratio
Subscribers
/city population
ratio
Free-floating car-
sharing*
Rome
4222631
2761632
824049
19.52%
29.84%
Milan
3237101
1371498
934777
28.88%
68.16%
Florence
994717
367150
164720
16.56%
44.86%
Turin
2205104
848885
266027
12.06%
31.34%
Bologna
1015701
392203
57546
5.67%
14.67%
Arezzo
334634
96672
229
0.07%
0.24%
Station-based car-
sharing**
Rome
4222631
2761632
3200
0.08%
0.12%
Turin
2205104
848885
12779
0.58%
1.51%
Genoa
816250
560688
2774
0.34%
0.49%
Brescia
1254322
196850
65
0.01%
0.03%
Bolzano
535774
107025
1142
0.21%
1.07%
Trento
542158
118509
700
0.13%
0.59%
Parma
450044
196655
695
0.15%
0.35%
Trapani
415233
64486
124
0.03%
0.19%
Palermo
1199626
630828
5133
0.43%
0.81%
Enna
155982
27586
47
0.03%
0.17%
Catania
1068835
298324
500
0.05%
0.17%
Cagliari
419770
148881
2016
0.48%
1.35%
3
The population of provinces where shared mobility service is available in Italy, as listed at
https://www.tuttitalia.it/province/ - accessed 22 October, 2022.
4
The population of cities where shared mobility service is available in Italy, as listed at
https://www.tuttitalia.it/citta/popolazione/ - accessed 22 October, 2022.
Free-floating Car-sharing,
238637 (1%)
Station-based Car-sharing,
6241149 (32%)
Free-floating bike-sharing,
2969412 (15%)
Station-based bike-sharing,
2778629 (14%)
Scooter-sharing,
7418938 (38%)
The number and percentage of rentals
Free-floating Car-sharing Station-based Car-sharing
Free-floating bike-sharing Station-based bike-sharing
Scooter-sharing

132
Type of shared
mobility service
City
Name
Province
population
City
population
Number of service
subscribers in 2020
Subscribers/province
population ratio
Subscribers
/city population
ratio
Free-floating
bike-sharing***
Rome
4222631
2761632
201564
4.77%
7.30%
Milan
3237101
1371498
442521
13.67%
32.27%
Station-based
bike-sharing****
Turin
2205104
848885
4159
0.19%
0.49%
La Spezia
214879
92216
876
0.41%
0.95%
Como
594657
83626
691
0.12%
0.83%
Bergamo
1102670
120207
370
0.03%
0.31%
Brescia
1254322
196850
29400
2.34%
14.94%
Bolzano
535774
107025
710
0.13%
0.66%
Trento
542158
118509
1434
0.26%
1.21%
Treviso
876755
84793
337
0.04%
0.40%
Padua
930898
209829
1004
0.11%
0.48%
Udine
517848
97761
1700
0.33%
1.74%
Trieste
230623
200594
10480
4.54%
5.22%
Parma
450044
196655
605
0.13%
0.31%
Modena
702787
185644
3188
0.45%
1.72%
Ravenna
386007
156080
2633
0.68%
1.69%
Forlì
391524
116861
242
0.06%
0.21%
Pisa
417245
89828
1145
0.27%
1.27%
Siena
262046
53724
254
0.10%
0.47%
Terni
218254
107314
37
0.02%
0.03%
Reggio
Calabria
518978
182455
511
0.10%
0.28%
Palermo
1199626
630828
3207
0.27%
0.51%
Scooter-
sharing*****
Rome
4222631
2761632
390734
9.25%
14.15%
Milan
3237101
1371498
204070
6.30%
14.88%
Turin
2205104
848885
118882
5.39%
14.00%
Bergamo
1102670
120207
4231
0.38%
3.52%
Monza
870112
122099
22503
2.59%
18.43%
Verona
927108
257274
50098
5.40%
19.47%
Parma
450044
196655
34873
7.75%
17.73%
Modena
702787
185644
7894
1.12%
4.25%
Rimini
336916
150051
50000
14.84%
33.32%
Pisa
417245
89828
8927
2.14%
9.94%
Pesaro
351993
94237
2042
0.58%
2.17%
Naples
2967117
914758
22666
0.76%
2.48%
Bari
1224756
316140
62457
5.10%
19.76%
Lecce
772276
95253
31263
4.05%
32.82%
* Free-floating car-sharing systems are also offered in Venice, Parma, Ferrara, Latina, Naples, Palermo, and Cagliari; however, since
the number of subscribers is unknown, they are not reported in the table.
** Station-based car-sharing systems are also offered in Milan, Venice, Padua, Arezzo, Messina, and Sassari; however, since the
number of subscribers is unknown, they are not reported in the table.
*** Free-floating bike-sharing systems are also offered in Turin, Bergamo, Mantua, Venice, Padua, Reggio Emilia, Bologna, Ferrara,
Florence, and Pesaro; however, since the number of subscribers is unknown, they are not reported in the table.
****Station-based bike-sharing systems are also offered in Genoa, Milan, Verona, and Livorno; however, since the number of
subscribers is unknown, they are not reported in the table.
***** Scooter-sharing systems are also offered in La Spezia, Trento, Venice, Ravenna, Cesena, Florence, Latina, Pescara, Caserta,
and Taranto; however, since the number of subscribers is unknown, they are not reported in the table.
5.1.2 Description of the study area and shared mobility services in Turin
The study area is located in the northwestern part of Italy. It includes the metropolitan area of
Turin, which consists of the municipality of Turin and its surrounding municipalities. In the
former, about 800,000 people live in about 130 square kilometers, while in the latter, about
544,000 people live in about 708 square kilometers. The population density in Turin is about
7,014 people per square kilometer and about 909 people per square kilometer outside the city
(Agenzia per la Mobilità Metropolitana e Regionale, 2015).
The motorization rate in metropolitan Turin is one of the highest in Italy, with around
664 private cars per 1000 inhabitants in 2017 (Regione Piemonte, 2017). In addition, most
residents of the Turin metropolitan area are satisfied with the various transportation services.

133
Specifically, in 2013, approximately 83% of the population was satisfied with public
transportation services, 88% with their car, and 92% with bikes (Agenzia per la Mobilità
Metropolitana e Regionale, 2015). Therefore, the diffusion of private cars and satisfaction with
public transportation and other active modes in the metropolitan area of Turin makes this study
area a good test bed for the analysis of the introduction of shared mobility services, as shared
transport modes were introduced where existing travel modes usage was consolidated.
The districts of Turin are the 8
5
administrative macro-zones into which the city of Turin
has been divided since 2016, with relative civic centers. In turn, the district group a total of 94
statistical zones divided into 34
6
corresponding city districts. Figure 17 depicts the name of
each district.
Figure 17: Map of the district of Turin
7
.
Furthermore, as demonstrated in Figure 18, each of the 31 municipalities surrounding Turin
corresponds to a specific zone (Agenzia per la Mobilità Metropolitana e Regionale, 2015).
5
Turin, Italy, has 8 administrative macro-zones, as mentioned on
https://www.museotorino.it/view/s/6de880fd1093417bbf1558809ff07266 - Accessed 22, September, 2021.
6
Turin, Italy, has 34 districts, as mentioned on
http://www.comune.torino.it/statistica/osservatorio/annuario/2002/pdf/03_Territorio.pdf - Accessed 22
September, 2021
7
A map of the 34 districts of Turin By .mau. at Italian Wikipedia, CC BY-SA 4.0,
https://commons.wikimedia.org/w/index.php?curid=63326088 - Accessed- 22 September, 2021.

134
Figure 18: Map of the Traffic Analysis Zones outside the municipality of Turin (Agenzia per
la Mobilità Metropolitana e Regionale, 2015).
According to Ciuffini et al. (2021), Turin is one of the few cities in Italy where the provision
of all three shared mobility services, including car-sharing, bike-sharing, and scooter-sharing,
is well-developed. As such, it is a good case study for that country. The number of station-
based car-sharing rentals in Turin in 2020 was 114128. Also, the number of Free-floating car-
sharing rentals in Turin in 2019, 2020, and 2021 were 1720224, 1002327, and 845323,
respectively. This drop in the number of Free-floating car-sharing rentals in Turin from 2019
to 2021 reflects the impact of Covid-19 on the use of car-sharing (Ciuffini et al., 2021; 2022).
In 2020, Turin had 278806 car-sharing subscribers (266027 for Free-floating and 12779 for
Station-based car-sharing). Besides, in Turin, the average distance traveled by car-sharing in
2020 was 6 km, and the average duration of its use was 27 minutes. In Turin, the total average
distance traveled by all car-sharing in 2020 was 6723588 km (5879041 km for Free-floating
car-sharing and 844547 km for Station-based car-sharing). Further, it should be stated that the
number of car-sharing fleets in Turin in 2020 was 881 (557 Free-floating car-sharing vehicles

135
and 324 Station-based car-sharing vehicles). Three car-sharing services
8
were in Turin: Enjoy
(Free-floating car-sharing), Car2go (Share Now) (Free-floating car-sharing), and BlueTorino
(Electric Station-based car-sharing) in 2021.
Regarding bike-sharing, it can be stated that two bike-sharing services
9
, ToBike
(Station-based bike-sharing) and Mobike (Free-floating bike-sharing), provided services in
Turn in 2021. In 2020, TOBike offered a fleet of 300 Station-based shared bikes, and the fleet
size of operator Movi's free-floating bike-share service was 1550. Also, the number of Station-
based bike-sharing rentals in Turin in 2020 was 159285. Additionally, in Turin, the total
average distance traveled by Station-based bike-sharing in 2020 was 476581 km. Moreover, in
2020, there were 4159 station-based bike-sharing subscribers.
Furthermore, in 2021, there were 3000 scooter-sharing fleets with six services in Turin.
In 2021, there are nine scooter-sharing services
10
, including Bird, BIT mobility, Dott, Helbiz
An, Circ, Lime, Wind, Link, and Vo i. In 2020, there were 1079032 rental scooter-sharing in
Turin. Besides, in Turin, the total average distance traveled by all scooter-sharing in 2020 was
1941837 km. Moreover, in 2020, there were 118882 scooter-sharing subscribers.
Furthermore, it is important to mention that from 2021 to 2022, some new shared
transportation services have been added, and some shared moving services have disappeared.
Turin has ten scooter-sharing services
11
in 2022, including Californian Bird, BIT mobility,
Bolt, Circ, Dott, Helbiz An, Lime, Link, Tier, and Vo i. Besides, regarding bike-sharing, it
should be mentioned that two operators
12
, ToBike and Ridemove operators, provide services
in 2022. Also, three operators
13
, LeasysGO, Enjoy, and ShareNow, offer services for car-
sharing in 2022.
5.2 Questionnaires design
In this study, nine different types of surveys are designed to understand the perspective of four
different main stakeholders (government members, operators, users, non-users) of the three
different shared mobility services (car-sharing, bike-sharing, scooter-sharing services) and
8
Turin, Italy, had three car-sharing services in 2021, as mentioned on
https://piemonte.movimentoconsumatori.it/news/car-sharing-e-sharing-mobility-a-torino-unalternativa-al-
trasporto-pubblico/ - accessed 22, November 2021.
9
Turin, Italy, had two bike-sharing services in 2021, as mentioned on
https://piemonte.movimentoconsumatori.it/news/car-sharing-e-sharing-mobility-a-torino-unalternativa-al-
trasporto-pubblico/ - accessed 22, November 2021.
10
Turin, Italy, had ten scooter-sharing services in Turin in 2021, as mentioned on
http://www.comune.torino.it/torinogiovani/vivere-a-torino/sharing-di-monopattini-elettrici-a-torino - accessed
22, November 2021.
11
Turin, Italy, has ten scooter-sharing services in 2022, as mentioned on
http://www.comune.torino.it/torinogiovani/vivere-a-torino/sharing-di-monopattini-elettrici-a-torino - accessed
20, September 2022.
12
Turin, Italy, has two bike-sharing services in 2022, as mentioned on
http://www.comune.torino.it/torinogiovani/vivere-a-torino/bike-sharing-e-noleggio-bici-a-torino - accessed
20, September 2022.
13
Four car-sharing operators offer services in Turin in 2022, as mentioned on
http://www.comune.torino.it/torinogiovani/vivere-a-torino/car-sharing-a-torino#carsharing – accessed 20,
September 2022.

136
shared mobility services (as a whole). It is important to note that the surveys are the same for
users and non-users. Also, government members and operators answered identical surveys for
each shared mobility service. The designed surveys are given in Appendix 2.
All stakeholders were asked to answer surveys and rank criteria. Although one or two
people in an organization generally make the final decisions, they obtain their information from
consultants who analyze and make recommendations. Hence, in some cases (if more were
available), more than one or two operators or government members have responded to the
surveys (for each shared mobility service).
In this study, nine different surveys are used to understand the perspectives of four
stakeholders of three shared mobility services, including car-sharing, bike-sharing, and
scooter-sharing (individually), as well as shared mobility services (as a whole, not for a specific
shared mobility service). These nine surveys, numbered from 1 to 9, are listed as follows.
• Survey 1: users and non-users of car-sharing services
• Survey 2: users and non-users of bike-sharing services
• Survey 3: users and non-users of scooter-sharing services
• Survey 4: government members and operators of car-sharing services
• Survey 5: government members and operators of bike-sharing services
• Survey 6: government members and operators of scooter-sharing services
• Survey 7: users and non-users of shared mobility services (as a whole, not for a
specific shared mobility service)
• Survey 8: government members who respond to the shared mobility services (as a
whole, not for a specific shared mobility service) surveys
• Survey 9: operators of shared mobility services (as a whole, not for a specific shared
mobility service).
Figure 19 shows these nine types of surveys (nine line arrows) associated with
stakeholders and each shared mobility service (car-sharing, bike-sharing, and scooter-sharing)
as well as shared mobility services (as a whole). In Figure 19, each line arrow drawn between
stakeholders and shared mobility services helps to understand which stakeholder is responding
to the survey associated with each shared mobility service. Users and non-users answer only
one survey among the car-sharing, bike-sharing, and scooter-sharing surveys (surveys 1, 2, 3,
or 7). Also, each government member responds to one of the car-sharing, bike-sharing, and
scooter-sharing surveys (surveys 4, 5, or 6), plus one survey associated with the shared mobility
service (as a whole, not for a specific shared mobility service) (survey 8). Besides, each
operator answers a survey related to the service operator (surveys 4, 5, or 6), plus answers one
survey associated with the shared mobility service (as a whole, not for a specific shared
mobility service) (survey 9).

137
Figure 19: Stakeholders and the survey associated with each shared mobility service to
which they responded.
Each survey can have different aspects according to the purpose for which it is designed. For a
better understanding, these aspects are given below.
• Question set A, BWM-related questions: these questions help to determine stakeholders'
views on the importance of the criteria and sub-criteria (if needed), such as cost and
travel time. An example of these questions (related to users/non-users of shared
mobility services (as a whole, not for a specific shared mobility service)) is given in
Figure 20, taken from survey 7.
• Question set B, Routines, daily travel views: these help to figure out the routines and
daily travel views of the users and non-users of each shared mobility service. For
instance, it contributes to knowing which mode of transportation users and non-users
are most likely to use to get to work or school. An example of these questions is given
in Figure 21, taken from surveys 1 to 3. It is also important to note that non-users are
not currently using the service (some have experience using it, and some have not);
hence, some of the questions are hypothetical concerning the use of the service.

138
• Question set C, Socio-demographic characteristics questions: they contribute to
understanding the socio-demographic characteristics of the users and non-users of each
shared mobility service, such as gender, age, and educational level. An example of these
questions is given in Figure 22, taken from surveys 1 to 3.
• Question set D, Characteristics that might induce non-users to use and also users to
use more shared services: this help to understand the views of government members
and operators of each shared mobility service on the characteristics such as departure
time and travel distance that might induce people to use (or use more). An example of
these questions is given in Figure 23, taken from surveys 4 to 6.
• Question set E, Characteristics affecting the use of shared mobility services: they help
to explore the perspectives of users/non-users of each shared mobility service (as a
whole, not for a specific shared mobility service) on some characteristics such as travel
speed and safety affecting the use of shared mobility services)). It will be used for the
Multi-Actor Multi-Criteria Analysis. An example of these questions is given in Figure
24, taken from survey 7.

139
Figure 20: Screenshot of the survey with BWM-related questions (question set A in survey
7).
Figure 21: Screenshot of the survey with routines and daily travel views questions (question
set B in surveys 1 to 3).

140
Figure 22: Screenshot of the survey with socio-demographic characteristics questions
(question set C in surveys 1,2 and 3).
Figure 23: Screenshot of the survey with questions about some characteristics that might
induce people to use (or use more) (question set D in surveys 4, 5, and 6).

141
Figure 24: Screenshot of the survey with questions about some characteristics affecting the use of
shared mobility services (question set E in survey 7).
To better understand the design of the nine surveys and their various aspects, first, section 5.2.1
explains the surveys associated with stakeholders of car-sharing, bike-sharing, and scooter-sharing
services (surveys 1 to 6). Then, section 5.2.2 describes the surveys associated with stakeholders of
shared mobility service services (as a whole, not for a specific shared mobility service) (surveys 7 to
9).
5.2.1 Surveys associated with stakeholders of car-sharing, bike-sharing, and
scooter-sharing services (surveys 1 to 6)
This section is dedicated to surveys associated with stakeholders (users, non-users,
operators, and government members) of car-sharing, bike-sharing, and scooter-sharing services
(surveys 1 to 6). In this section, it is important to note that surveys 1 to 3 for users and non-
users were similar. Also, government members and operators answered identical surveys
(surveys 4 to 6) for each shared mobility service. For the four stakeholders, the BWM-related
questions (question set A in surveys 1 to 6) were the same (to understand the difference in their
views on the same factors). Still, the rest of the questions users and non-users (surveys 1 to 3)
asked differed from those of government members and operators (surveys 4 to 6). In subsection
A2.1 and A2.2 (in Appendix 2), two surveys are presented separately for users and non-users
stakeholders (surveys 1 to 3) and the government members and operators stakeholders (surveys
4 to 6), respectively. The explanation of these surveys is as follows.

142
• There are questionnaires for users and non-users of each shared mobility service. This
type of survey is designed for users and non-users of car-sharing, bike-sharing, and
scooter-sharing services, and it includes two parts (surveys 1 to 3). In the first part, there
are questions related to BWM analysis (question set A in surveys 1 to 3). In the second
part, there are questions relevant to the respondents’ routines, daily travel views
(question set B in surveys 1 to 3), and socio-economic situation (question set C in
surveys 1 to 3). Hence, in addition to BWM-related questions (question set A in surveys
1 to 3), questions about their routines, daily travel views (question set B in surveys 1 to
3), and socio-demographic characteristics (question set C in surveys 1 to 3) are also
included in the surveys (surveys 1 to 3), most of which were taken from the STARS
project questionnaire
14
. This helps to have standard and precise questions in the surveys
(surveys 1 to 3).
There are questionnaires for government members and operators of each shared
mobility service. This type of survey is designed for government members and
operators of car-sharing, bike-sharing, and scooter-sharing services (surveys 4 to 6),
and it includes two parts. In the first part, there are questions related to BWM analysis
(question set A in surveys 4 to 6). In the second part, questions are relevant to the
respondent’s opinion about some of the characteristics that might induce people to use
(or use more) {car, bike, scooter}-sharing (question set D in surveys 4 to 6).
5.2.2 Surveys associated with stakeholders of shared mobility service
services (as a whole, not for a specific shared mobility service) (surveys 7 to
9)
This section is dedicated to surveys associated with stakeholders (users, non-users, operators,
and government members) of shared mobility services (as a whole, not for a specific shared
mobility service), i.e., surveys 7 to 9 in the above list. In this section, it should be noted that
the type of surveys conducted among government members (survey 8) and operators (survey
9) for shared transportation services (as a whole, not for a specific shared mobility service) was
different because the purpose was to understand the importance of factors related to their
decision, which were different for these two groups. Hence, three surveys for users/non-users
(survey 7), government members (survey 8), and operators (survey 9) of shared mobility
services (as a whole, not for a specific shared mobility service) are presented separately in the
three subsections A2.3, A2.4, and A2.5, respectively. The description of these surveys (7 to 9)
is as follows.
• There are questionnaires for users and non-users of shared mobility services (as a
whole, not for a specific shared mobility service) (survey 7). This type of survey is
14
STARS project was Launched in October 2017. This project aimed to investigate the diffusion of car-sharing
in Europe, its relationships with technological and social innovations, and its effect on other transport modes
such as bicycles, walking, cars, public transport, and taxis.
Questions about people's routines, daily travel views, and socio-demographic characteristics in the surveys were
taken from the STARS project questionnaires available on
https://zenodo.org/record/3608887#.YswGVnZBy3B - accessed 11 November 2021.

143
designed for users and non-users of shared mobility services (as a whole, not for a
specific shared mobility service), and it includes two parts. In the first part, there are
questions related to BWM analysis (question set A in survey 7). In the second part,
questions are relevant to the respondent’s opinions on characteristics affecting car-
sharing, bike-sharing, and scooter-sharing use (question set E in survey 7).
• There is a questionnaire for government members about shared mobility services (as a
whole, not for a specific shared mobility service) (survey 8). This type of survey is
designed for government members and is about shared mobility services (as a whole,
not for a specific shared mobility service). In this survey, there are questions related to
BWM analysis (question set A in survey 8).
• There is a questionnaire for operators of shared mobility services (as a whole, not for a
specific shared mobility service). This type of survey is designed for operators of shared
mobility services (as a whole, not for a specific shared mobility service). In this survey,
there are questions related to BWM analysis (question set A in survey 9).
The above importance ranking exercise was complemented by 3*8 = 24 rating questions to
gather the respondents’ evaluations on the performance of car-sharing, bike-sharing, and
scooter-sharing related to each of these eight criteria. The 7-point semantic scales were used
(question set E in survey 7) to this effect, ranging, for instance, from very unsafe to very safe
for the first criterion, from very poor to very good for the second criterion, and so on. An
example of a 7-point semantic scale question is illustrated in Figure 25.

144
Figure 25: A sample of a 7-point semantic scale question (question set B in survey 7).
5.3 Data collection activities
In this study, SWG
15
collected data from 19/11/2021 to 09/ 02/2022. The data on operators and
government members were collected through phone calls (in Italian) to targeted contact points,
whereas for users and non-users, it was possible to resort to their panel to have a representative
sample of the population in the study area.
As the number of operators and government members was relatively small, data
collection was done by phone call to clarify the questions better (compared to an online survey)
and to obtain more accurate responses (surveys 4, 5, 6, 8, and 9). Furthermore, online surveys
have been used to collect data from users and non-users of car-sharing (survey 1), bike-sharing
(survey 2), scooter-sharing (survey 3), and shared mobility services (as a whole) (survey 7), as
it is standard practice with panels maintained by surveying companies. It is important to note
that although face-to-face data collection with individuals and clarifying questions could
provide better (less biased) answers, it was not possible to do it in person due to the relatively
large number of users and non-users. It should be mentioned that before data collection, all
online surveys were repeatedly reviewed by the author to ensure their accuracy and the absence
of problems in the online data collection process.
As an example, Figure 26 shows question 1 (B1) of the BWM online survey questions
(question set A in survey 1). The questions in the surveys are in Italian.
15
SWG, founded in 1981 in Trieste, is a leading Italian company in surveys, market research, sector studies, and
observatories (https://www.swg.it/).

145
Figure 26: Screenshot from the original online survey (first BWM question (question B1))
(question set A in survey 1).
The survey data is utilized to calculate the criteria and sub-criteria weights to determine
how the comparative criteria are rated in terms of importance by different stakeholders of
different shared mobility services. Hence, surveys help to gain insights into how specific
individuals or groups perceive specific aspects. In addition, it contributes to constructing
criteria/sub-criteria weights and assists in understanding how the weights receive scores
(against each other).
5.4 Collected data
To better understand the collected data, the number of stakeholders of each shared mobility
service to participate in surveys (surveys 1 to 9) (that was requested to SWG by the author) and
the number of stakeholders of each shared mobility service that responded to the surveys
(survey 1 to 9) are presented in Table 38.
Table 38: The number of survey responses requested (to SWG) and received from the
stakeholders of each shared mobility service (surveys 1 to 9).
Type of shared mobility
service
Stakeholders of shared mobility services
Government members
Operators
Users
Non-users
Requested
Received
Requested
Received
Requested
Received
Requested
Received
Car-sharing
3
4
3
3
15
76
15
126
Bike-sharing
3
5
3
3
15
75
15
127
Scooter-sharing
3
3
3
3
15
77
15
126
Shared mobility (as a
whole)
9
9
9
9
15
100
15
104
As seen in Table 38, the minimum number of survey responses requested to SWG is the same
among the same stakeholder of car-sharing, bike-sharing, and scooter-sharing services so that
the results can be better compared (surveys 1 to 6). Also, each government member was
supposed to respond to the shared mobility services (as a whole) survey (surveys 8). Hence, at
least nine shared mobility services (as a whole) surveys (survey 8) needed to be completed.
Operators are supposed to do the same (survey 9). However, it is important to mention that
since some of these responses to the BWM-related questions (set A in surveys 1 to 9) could be
omitted, the number of surveys administered was equal to or greater than the requested number,
especially for the user and non-user surveys (surveys 1, 2, 3, and 7). Besides, as the author

146
requested from SWG, the number of responses received from the same stakeholder (e.g.,
operators) is the same or similar for car-sharing, bike-sharing, and scooter-sharing services.
5.4.1 Socio-demographic characteristics of users and non-users
It is essential to mention that this study assumes that the probability of being part of the survey
panel is completely unrelated to the probability of being a shared mobility subscriber. With this
assumption, it can be claimed that the results are valid for the general population (all users and
non-users in Turin). Additionally, the possible responses to the closed-form survey questions
used in this study are the same as the STARS project surveys. Therefore, most of the socio-
demographic ranges (question set C in surveys 1 to 3) used are the same socio-demographic
range used in the STARS project surveys (to be the standard ranges). The socio-demographic
characteristics of survey respondents who are users and non-users of car-sharing, bike-sharing,
and scooter-sharing services (question set C in surveys 1 to 3) are given in Table A10 in section
A4.1 of Appendix 4. Figures 27 and 28 present the percentage (as well as the absolute number)
of users and non-users of each shared mobility service (question set C in surveys 1,2, and 3
respondents), respectively, living in Turin and outside Turin. As offered in Figures 27 and 28,
the majority of users and non-users of car-sharing, bike-sharing, and scooter-sharing services
(question set C in surveys 1, 2, and 3 respondents) live in Turin, which is the case study of this
research.

147
Figure 27: Percentage (as well as the absolute number) of users of each shared mobility
service (question set C in surveys 1 to 3 respondents) living in Turin and outside Turin.
Figure 28: Percentage (as well as the absolute number) of non-users of each shared mobility service
(question set C in surveys 1 to 3 respondents) living in Turin and outside Turin.
5.4.2. Routines and daily travel views of users and non-users
The routines and daily travel views of survey respondents who are users and non-users of car-
sharing, bike-sharing, and scooter-sharing services (question set B in surveys 1 to 3) are given
in Table A11 in section A4.2 of Appendix 4. The percentage (as well as the absolute number)
77.63% 85.33%
76.62%
22.37%
14.67% 23.38%
59
64 59
17 11 18
0
10
20
30
40
50
60
70
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Car-sharing Bike-sharing Scooter-sharing
Number of users
Percentage of users
Type of shared mobility service
Users of each shared mobility service
Percentage of users in Turin Percentage of users outside of Turin
Number of users in Turin Number of users outside of Turin
59.52% 72.44%
76.62%
40.48% 27.56%
23.38%
75 92 82
51 35 44
0
10
20
30
40
50
60
70
80
90
100
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Car-sharing Bike-sharing Scooter-sharing
Number of non-users
Percentage of non-users
Type of shared mobility service
Non-users of each shared mobility service
Percentage of non-users in Turin Percentage of non-users outside of Turin
Number of non-users in Turin Number of non-users outside of Turin

148
of users and non-users of each shared mobility service who use and do not use their private car
daily is shown in Figures 29 and Figure 30, respectively. Also, it should be mentioned that
nearly 40% of non-users of bike-sharing users use their private cars daily; however, this figure
for bike-sharing users is only 16%. On the other hand, almost 30% of bike-sharing users use it
1-3 days a week; however, this figure for non-users is about 17%. Hence, unlike bike-sharing,
car-sharing and scooter-sharing usage do not remarkably impact reducing private car use.
n
Figure 29: The percentage (as well as the absolute number) of users of each shared mobility
service who use and do not use their private car on a daily basis (question set B in surveys 1
to 3 respondents).
32.89%
16.67%
34.72%
67.11%
83.33%
65.28%
25
12
25
51
60
47
0
10
20
30
40
50
60
70
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Car-sharing Bike-sharing Scooter-sharing
Number of users
Percentage of users
Type of shared mobility service
Users of each shared mobility service
Percentage of users who use private cars daily
Percentage of users who do not use private cars daily
Number of users who use private cars daily
Number of users who do not use private cars daily

149
Figure 30: The percentage (as well as the absolute number) of non-users of each shared
mobility service who use and do not use their private car on a daily basis (question set B in
surveys 1 to 3 respondents).
The percentage of car-sharing and bike-sharing users who have pick-up locations near their
home (or their home is in an operational area) is at least 1.5 times higher than that of non-users
(who are at least familiar with the service). This numerical ratio is similar between car-sharing
and bike-sharing users and non-users for whom pick-up locations are close to their most
frequent destinations (or locations are in the operational area). Therefore, the presence of pick-
up locations near home and the destination area can increase the demand for car-sharing and
bike-sharing among people who are at least familiar with the service.
Approximately 34% of car-sharing users use car-sharing once a few times a month.
Interestingly, around 59% of bike-sharing and 62% of scooter-sharing users rarely or never use
car-sharing. It is interesting to know that the use of public transport is higher among scooter-
sharing users than non-users. It represents the integration of scooter-sharing services with
public transportation systems. Besides, it can point out that most car-sharing, bike-sharing, and
scooter-sharing users and non-users are reluctant to use motorcycles/scooters, taxis, and
personal bikes; however, both are interested in daily walking. Also, most car-sharing and bike-
sharing users choose public transport and walking to do an errand in the city center, while most
scooter-sharing users and the majority of the three shared mobility non-users use their car (as
a driver) for this purpose. In addition, most of both car-sharing, bike-sharing, and scooter-
sharing users and non-users prefer their private car (as a driver) for other trip purposes,
including going to work or school, visiting a close relative/friends/relatives/family, going out
for dinner, taking an excursion in nice weather, visiting a shopping center, and for weekend
35.59%
39.66%
32.20%
64.41% 60.34%
67.80%
42
46 38
76
70
80
0
10
20
30
40
50
60
70
80
90
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Car-sharing Bike-sharing Scooter-sharing
Number of non-users
Percentage of non-users
Type of shared mobility service
Non-users of each shared mobility service
Percentage of non-users who do not use private cars
daily
Percentage of non-users who use private cars daily
Number of non-users who use private cars daily
Number of non-users who do not use private cars
daily

150
activities. Also, users and non-users of these shared mobility services prefer walking to other
transport modes to go to smaller shops. It is also worth stating that the highest percentage of
individuals likely to use car-sharing for travel purposes (among those mentioned in Table A11
in section A4.2 of Appendix 4) is about 15.8%, which is to perform a work-related activity in
the city center. However, the highest percentage of people likely to use bike-sharing and
scooter-sharing for travel purposes (among those listed in Table A11 in section A4.2 of
Appendix 4) is around 10.7% and 6.5%, respectively, related to weekend activities.
Interestingly, both users and non-users groups believe that the impact of health concerns
caused by the Covid-19 pandemic does not reduce their motivation to use. Also, it is found that
the majority of users and non-users are of the opinion that cost reduction is the most important
factor (among the factors asked) that may encourage them to use (or use more) car-sharing.
Therefore, it can indicate the importance of the effect of a service cost on demand. Further, it
should be noted that the availability of a scooter-sharing service close to home/work is the
biggest motivation for using this service for both users and non-users. This indicates the high
impact of the availability of scooter-sharing on its demand. Likewise, the availability of bike-
sharing is the most important motivation for users to use bike-sharing. However, in the eyes of
non-users of bike-sharing, the most important reason that can encourage them to use bike-
sharing is the convenience of having it only when needed. It is interesting to know that a smooth
and non-sloping path does not greatly affect the use of shared bikes and scooters.
Moreover, bad weather (e.g., rainy or snowy) is the most important weather condition
that can drive most users and non-users to use car-sharing. This shows the important role that
car-sharing can play as a mode of transportation in inclement weather. On the other hand, for
most users and non-users of bike-sharing and scooter-sharing, good weather (e.g., sunny
weather) is the weather condition that induces them to use the service (answers only belong to
people who are at least familiar with the service). It should also be mentioned that the humidity
and air pollution levels do not affect the demand for these three types of shared transportation
services. In addition, among the seasons, winter is the season when most car-sharing users use
this service, whereas most bike-sharing and scooter-sharing users use the service in spring.
The distance that may persuade most users and non-users of bike-sharing and scooter-
sharing services to use the service is less than 5 km. Also, most users and non-users of car-
sharing, bike-sharing, and scooter-sharing services prefer to use the service for less than 30
minutes, demonstrating the importance of shared mobility services for short trips. Although
most car-sharing users prefer to use this service during off-peak hours, most bike-sharing and
scooter-sharing users and non-users prefer to utilize this service during peak hours. This shows
the important role each shared mobility service can play in the transportation system at certain
times.
Furthermore, weekday morning is chosen as the preferred departure time by the
majority of users and non-users of each shared mobility service (they could select more than
one departure time option in their preference in the survey). This indicates that the departure
time that might cause them to use the shared mobility service is a weekday morning. This
reveals the undeniable role of shared transportation services in weekday morning transportation
systems.

151
Interestingly, around 41% of non-users would like to use car-sharing for non-leisure
trips (going to school or work); however, about 39.5% of car-sharing users choose this service
for leisure trips (e.g., visiting friends or shopping) and non-leisure trips. This shows that non-
users are not paying attention to the potentialities of this service for leisure travel. Regarding
bike-sharing services, bike-sharing could be used for leisure trips by approximately 43% of
non-users (if they want to use it); however, almost 37% of bike-sharing users use this service
for leisure and non-leisure trips. This demonstrates that non-users have not considered the
capacity of this service for non-recreational trips. Finally, in regard to scooter-sharing services,
about 40% of scooter-sharing users prefer to use this service for travel for leisure trips, while
almost 44% of non-users would use this service for non-leisure trips. This shows that this
service has the potential to be used for both travel purposes.
It should be remarked that car-sharing, bike-sharing, and scooter-sharing are relatively
enjoyable for users and for non-users (who have experience using the service but no longer use
it). This suggests that car-sharing is enjoyable for this kind of non-users; hence, there are other
reasons behind not using the service. It should be stated that about 40% of non-users of bike-
sharing (who have previous experience) disagree that bike-sharing provides a good service.
This might be one of the reasons why they do not use it anymore. Moreover, they are less likely
than users to agree that the service is predictable and trustworthy. These are other reasons that
make them less attracted to the service.
It is important to note that most of the non-users of car-sharing, bike-sharing, and
scooter-sharing (who do not have the experience of using the service but are familiar with it)
do not support its implementation well in society, especially compared to users. This is because
their view of the service is less favorable compared to users. It is worth mentioning that most
of these non-users disagree that they are sure they can choose this service for their regular trips
in the next week. Therefore, their decision not to use the service is profound. Interestingly, they
disagree that booking on the website/app is complex. Therefore, these services are user-
friendly; hence, there is no need to invest much in this sector (making the service more user-
friendly) to attract this kind of non-users to use these services.
Furthermore, the majority of both groups agree that using these shared mobility services
is relatively environmentally friendly. Besides, users and non-users believe that the urgent need
to reduce ecological destruction caused by car use has not been overestimated. Also, they
believe that the use of cars brings many environmental problems. Further, non-users, like users,
would feel better if they traveled more sustainably. Therefore, awareness of environmental
issues and interest in reducing related problems is not sufficient motivation for non-users to
use shared mobility services. It can also be pointed out that the political orientation of users
and non-users is neither left nor right. Therefore, people's political orientation does not affect
their use or non-use of shared transportation services.
It is essential to state that differences in the routines and daily travel patterns of male
and female users of each shared transportation service (question sets B and C in surveys 1 to
3) can be seen as shown in Table A12 in section A4.2 of Appendix 4. These differences include
the motivation to use the service, the time of departure, and the purpose of the trip that may
cause the use of the service. As delivered in Table A12, the role of cost reduction as a

152
motivation to use car-sharing is greater for males than females. On the other hand, increased
comfort during travel (by car-sharing) is more important for females than males as a motivation
to use the service. In the case of bike-sharing users, it is interesting to note that increased
comfort during travel is more important to male bike-sharing users than female users. On the
other hand, it can be pointed out that the availability of the service near the user's home/work
and avoiding responsibilities related to maintenance and repairs are more important for females
than males as a motivation to use bike-sharing. Regarding the incentives that may make users
use scooter-sharing, it should be noted that more sustainable travel and increased comfort
during travel (by using scooter-sharing) are more critical motivations for males than females.
On the other hand, reducing costs and avoiding responsibilities related to maintenance and
repairs are more important for females than males.
Interestingly, 31% and 25.86 % of times, the weekday evening and weekend morning,
respectively, were chosen by male car-sharing users as preferred departure times. Meanwhile,
31.58% and 29.82% of times, the weekend evening and weekday morning, respectively, are
chosen by female car-sharing users as the preferred departure time.
It can also be noted that compared to female car-sharing users, male car-sharing users
are more interested in using the service only for non-leisure (going to work/school) trips.
Meanwhile, in comparison to female bike-sharing users, male bike-sharing users are more
inclined to use the service only for leisure (e.g., visiting friends or shopping) trips. Also,
regarding traveling only for non-leisure (going to work/school), female bike-sharing users are
more interested than male bike-sharing users. Furthermore, concerning leisure-only travel (e.g.,
visiting friends or shopping), female scooter-sharing users are keener than male users.
5.4.3 Selected data (responses to the BWM-related questions) in this study
First, it should be noted that the members of the government and executives have
acknowledged that they agree with the criteria and sub-criteria used in this research (according
to the goals of this research) and have not added a new one (before responding to the BWM-
related questions (question set A in surveys 8 and 9)).
To be more familiar with the selected data that passed the quality check, unacceptable
data (responses to the BWM questions (set A in surveys 1 to 9)) should be excluded. In this
regard, before calculating the optimal group weights by Bayesian BWM, one can check the
global input-based consistency ratio obtained using Equations 27 and 28 (in section 3.2.7.5.1.2
of Chapter 3). Before calculating the optimal group weights by Bayesian BWM, the
consistency of the respondents can be examined using the Input-based approach (Eq. (27 and
28) in section 3.2.7.5.1.2), and acceptable ones (their obtained global input-based consistency
ratio is less than the input-based consistency ratio thresholds) can be considered (Liang et al.,
2020). As mentioned in section 3.2.7.5.1.2 of Chapter 3, one of the advantages of using the
input-based approach is to obtain an immediate input-based consistency ratio to check (with its
thresholds). The response could be revised if the input-based consistency ratio was greater than
its thresholds. However, since face-to-face interaction with respondents was not possible in
this study and surveys and telephone calls were used for data collection by the SWG (not by

153
the author), this positive aspect of the input-based approach was not used. Hence, after
eliminating pairwise comparisons with unacceptable consistency ratios (section 3.2.7.5.1.2),
different sample sizes can be obtained and utilized for different levels of the model.
For more information about government members (surveys 4, 5, 6, and 8) and operators
(surveys 4, 5, 6, and 9) participating in the respective surveys whose responses to the BWM
questions (question set A in surveys 4, 5, 6, 8, and 9) were selected in this study, their job status
according to the type of shared transportation service is given in Table A2 to A9 in section
A3.6 of appendix 3.
As shown in Table 38, 76 respondents completed the survey on behalf of car-sharing
users (survey 1). However, not all these observations can be used for the Bayesian BWM
model. In fact, before calculating the optimal group weights, the consistency of the respondents
was also checked, and the ones with an acceptable consistency ratio were considered (Liang et
al., 2020). As a result, a different sample size was utilized for each set of criteria. A sample
size of 15 respondents (n=15) was used for the main-criteria set, a sample size of 39
respondents was used for the trip-related characteristics sub-criteria set (n=39), and a sample
size of 36 was used for the car-sharing characteristics sub-criteria set (n=36). For the
availability and accessibility sub-criteria-set, a sample size of 39 instead of 76 was used (n=39)
to obtain more reliable results.
Since there are only two criteria in the availability and accessibility sub-criteria for user
respondents, the mistake of not assigning the highest value to the best-worst vector does not
occur. As a result, all 76 respondents are only included in this subset because, technically, this
mistake cannot happen if there are only two criteria (best and worst). However, suppose this
subset contains more than two criteria. In that case, there could also be the possibility of
conducting the wrong pairwise comparison, leading to the omission of respondents in this set
(as is the case of the main set, trip-related characteristics subset, and car-sharing characteristics
subset). Therefore, the result may be less reliable because the data in this sub-criteria set is
based only on technicality. Hence, the second-highest sample size (closest to n=76) is
considered to determine the criteria weights for the availability and accessibility subset,
including respondents who performed the pairwise comparison correctly. As a result, the
respondents of the trip-related characteristics sub-criteria-set were used (n=39). This process
is done for other stakeholders as well. After these quality checks, the number of utilized
responses to the BWM questions of each stakeholder of car-sharing services (question set A in
surveys 1 and 4) for the main-criteria and each sub-criteria set is listed in Table 39.
Table 39: Number of responses that passed quality checks from each stakeholder for the
main-criteria and each sub-criteria set for the car-sharing, out of the total number of
responses shown in the last column (question set A in surveys 1 and 4).
Type of
stakeholder
Main criteria set (trip-
related characteristics,
car-sharing
characteristics, and
availability and
accessibility)
Sub-criteria Sets
Total
sample
size
Trip-related
characteristics (travel
time, travel distance,
departure time, trip
purpose)
Car-sharing
characteristics (cost,
comfort, safety, service
quality, environment-
friendly system, user-
friendliness)
Availability and
accessibility (service
availability, vehicle
availability, and
accessibility)
Users
15
39
36
39
76
Non-users
24
59
56
59
126

154
Operators
2
3
3
3
3
Government
Members
2
3
4
4
4
Similarly, the number of utilized responses to the BWM-related questions of each
stakeholder for the main criteria and each sub-criterion set for bike-sharing (question set A in
surveys 2 and 5) is listed in Table 40.
Table 40: The number of used responses from each stakeholder for the main-criteria and
each sub-criteria set for the bike-sharing (question set A in surveys 2 and 5).
Type of
Stakeholder
Main criteria set (trip-
related characteristics,
bike-sharing
characteristics, and
availability and
accessibility)
Sub-criteria Sets
Total
sample
size
Trip-related
characteristics (travel
time, travel distance,
departure time, trip
purpose)
Bike-sharing
characteristics (cost,
comfort, safety, service
quality, environment-
friendly system, user-
friendliness)
Availability and
accessibility (service
availability, vehicle
availability, and
accessibility)
Users
18
38
37
38
75
Non-users
32
69
63
69
127
Operates
2
3
3
3
3
Government
Members
2
4
4
4
5
The number of utilized responses to the BWM questions of each stakeholder for the main
criteria and each sub-criterion set for scooter-sharing (question set A in surveys 3 and 6) is
listed in Table 41.
Table 41: The number of used responses from each stakeholder for the main-criteria and
each sub-criteria set for the scooter-sharing (question set A in surveys 3 and 6).
Type of
Stakeholder
Main criteria set (trip-
related characteristics,
scooter-sharing
characteristics, and
availability and
accessibility)
Sub-criteria Sets
Total
sample
size
Trip-related
characteristics (travel
time, travel distance,
departure time, trip
purpose)
Scooter-sharing
characteristics (cost,
comfort, safety, service
quality, environment-
friendly system, user-
friendliness)
Availability and
accessibility (service
availability, vehicle
availability, and
accessibility)
Users
13
42
37
42
77
Non-users
24
66
48
66
126
Operates
1
3
3
3
3
Government
Members
2
3
3
3
3
Finally, the number of utilized responses to the BWM questions of each stakeholder of the
shared mobility services (as a whole) (question set A in surveys 7 to 9) is listed in Table 42.
Table 42: The number of used responses from each stakeholder of the shared mobility
services (as a whole) (question set A in surveys 7 to 9).
Type of Stakeholder
Number of used responses to the BWM questions for the criteria set
Users
45
Non-users
55
Operates
8
Government members
7

155
5.4.4. Socio-demographic characteristics of selected users and non-users of
each of the shared mobility services
The socio-demographic characteristics of survey respondents who are users and non-users of
car-sharing, bike-sharing, and scooter-sharing services (question set C in surveys 1 to 3), and
their responses to the BWM questions (question set A in surveys 1 to 3), which have been
utilized are mentioned in Tables A13 to A18, respectively, in section A4.3 of appendix 4.
5.4.5. Views of whole operators and members of the government regarding
some of the travel routines of users of each of the shared transportation
services
It is essential to know the opinions of operators (related to each shared mobility service) and
government members about some of the travel routines of users of each shared mobility service
(question set D in surveys 4 to 6), listed in Table A19 (in section 4.4 of Appendix 4). This
contributes to determining the gaps between the opinions of operators and government
members about the travel routine of users of each shared mobility and what users expressed
about it.
In this regard, it is important to mention that from the perspective of 56.58% of car-
sharing users (shown in Table A11 in section A4.2 of Appendix 4) and 50% (listed in Table
A11 in section A4.2 of Appendix 4) of government members (who responded to the car-sharing
survey) (survey 4), short-time trips (less than 30 min) can induce people to use (or use more)
car-sharing, however, trips beyond 30 min cannot do that. Furthermore, table A19 (in section
4.4 of Appendix 4) shows that none of the car-sharing operators agree with the statement. This
designates the gap between the views of car-sharing operators (question set D in survey 4) and
the perspective of car-sharing users (question set E in survey 1) and government members (who
responded to the car-sharing survey) (question set D in survey 4) about the effect of short-time
trips on car-sharing demand.

156
Chapter 6
Results
In this section, the results are offered. In this regard, after the problem definition and
alternatives selection (step 1 of MAMCA, mentioned in section 3.1.1 of Chapter 3 and 4.1 of
chapter 4), stakeholder analysis (step 2 of MAMCA, given in section 3.1.2 of Chapter 3 and
section 4.2 of Chapter 4), selection of the criteria for each study purpose (step 3 of MAMCA,
presented in section 3.1.3 of Chapter 3 and section 4.3 of chapter 4) and gathering the required
data (offered in Chapter 5), in order to obtain the weights (step 3 of MAMCA, presented in
section 3.1.3 of Chapter 3 and section 4.3 of chapter 4), first, the input-based approach is used
to eliminate the unacceptable responses (mentioned in section 3.2.7.5.1.2 of Chapter 3 and
section 5.4.3 of Chapter 5). Then, Bayesian BWM is used to find the weights of the criteria
(explained in section 3.2.7.6 of chapter 3). In this regard, more details on the analysis for each
shared mobility service (separately) and the analysis for shared mobility services (as a whole,
not for a specific shared mobility service) are presented in sections 6.1 and 6.2, respectively.
Finally, since the indicator (value) (clarified in section 3 of Chapter 3, and is step 4 of
MAMCA, explained in section 3.1.4 of Chapter 3) for each criterion associated with the users
and non-users of shared mobility services (as a whole, not for a specific shared mobility
service) is gathered, the perception-based analysis and sensitivity analysis and scenarios can
be given in sections 6.2.6 and 6.2.7 (clarified in section 3 of Chapter 3 and is in step 5 to 7 of
MAMCA, described in section 3.1.5 to 3.1.7 of Chapter 3).
6.1 Results of the Analysis for Each Shared Mobility Service
(Separately)
In this section, initially, under one specific shared mobility service, four groups of stakeholders
are compared in terms of their perception of a particular main-criterion/sub-criterion (four
different stakeholders have reviewed common criteria and sub-criteria). This contributes to
understanding how the perceptions of different groups of each shared mobility transport modes

157
stakeholders can be different about a specific main-criterion/sub-criterion (related to the first
research question mentioned in Chapter 1).
Furthermore, the differences in the importance of one main-criterion/sub-criterion
across the three types of shared mobility services, including car-sharing, bike-sharing, and
scooter-sharing, are examined for each specific type of stakeholder. This helps to understand
how one main-criterion/sub-criterion can be of different importance across different shared
mobility services (related to the second research question mentioned in Chapter 1).
Furthermore, suggestions can be given to operators and government members to show how the
importance of sub-criteria and main-criteria can be utilized to grow the users' engagement and
increase the attraction of non-users to services (related to the first research question mentioned
in Chapter 1). It is important to mention that these research questions mentioned in Chapter 1
can be answered with visual data (especially credal ranking of the criteria) and tables. However,
since Bayesian BWM is used in this study, the p-values of the comparison analysis method are
not applicable.
Since there are three shared mobility services, three sections (one section for each
shared mobility service) will be provided. Also, a section will be given to determine the
importance of each main-criterion/sub-criterion across the three types of shared mobility
services. Therefore, a total of four sub-sections for the results of this section are given below.
6.1.1 Car-sharing services
In this part, the group weight of each stakeholder, including government members, operators,
users, and non-users of car-sharing services, is analyzed for the main-criteria and sub-criteria.
This helps show their priority for the main-criteria and the sub-criteria.
6.1.1.1 Group weight of government members for car-sharing services
The optimal government members’ group weights of the main-criteria for car-sharing services
are listed in Table 43.
Table 43: Government members’ group weights of the main-criteria for car-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.3603
C2. Car-sharing characteristics
0.3824
C3. Availability and accessibility
0.2574
As presented in Table 43, from the point of view of government members, the most important
main-criterion that individuals could consider in using car-sharing is car-sharing characteristics
(C2), with a weight  = 0.3824. This means that government members believe that people
place more value on main criterion C2 when using car-sharing than on main-criteria trip-related
characteristics (C1) and availability and accessibility (C3). Figure 31 shows the credal ranking
of the main-criteria from the perspective of government members (for car-sharing services) and
the assigned confidence level CL. The definition of CL is given in section 3.2.7.6.4 of Chapter
3.

158
In the Bayesian BWM, the criteria can be compared through credal ranking graphs,
where the nodes are the criteria (e.g., C1, C2, and C3 in Figure 31). Also, on each edge, A 

B (e.g., C2 
󰇒

󰇏
C3 in Figure 31) indicates that criterion A is more important with confidence d
(degree of certainty about the relation of criteria) than B. The notation “confidence d” was
present in the main article (Mohammadi and Rezaei, 2020), in which the Bayesian Best-Worst
Method was introduced. However, the same value is also called the confidence level (CL) in
recent literature (Kalpoe, 2020). In this study, the latter notation is used. To be more precise,
in Bayesian BWM, confidence is basically the extent to which we can claim one criterion is
more important than the other. This comes from the probabilistic nature of the model.
The different colors indicate the relationship between each criterion and the less
important criteria. For example, in Figure 31, red is used for the relationship between C2 and
the less important criteria than C2 (C1 and C3). Also, blue is used for the relationship between
C1 and the less important criterion than C1 (C3).
Figure 31: Credal ranking of main-criteria from government members’ view for car-sharing
services.
Figure 31 shows that the main-criterion car-sharing characteristics (C2) has a relatively high
CL of 0.71 compared to the main-criterion availability and accessibility (C3). As mentioned in
section 3.2.7.6.4 of Chapter 3, when the threshold value is 50 and the CL is around 60 to 80, it
can be pointed out that one criterion is more important than the other. On the other hand, when
the threshold value is 50, and the CL is 50 (equal to the threshold value), or slightly higher
(from 50 to less than 60), the superiority of one criterion over another is not well established.
In this regard, the main-criterion C2 does not have a high CL compared to the main-criterion
trip-related characteristics (C1) (CL=0.53). In other words, the superiority (i.e., a more
important and influential factor in people's car-sharing use) of the main-criterion C2 over the
main-criterion C1 is not well established. Hence, although C2 is considered more important
than the other two main-criteria, a confidence of 0.53 between it and C1 implies that some

159
government members believe that C1 plays a more important role. On the other hand, between
C1 and C3, the former is more important than the latter, with a confidence of 0.68.
Table 44 presents the optimal group weights of government members for car-sharing
services. The main-criteria followed by the sub-criteria are listed. Also, the optimal groups’
local weights for each sub-criterion and the relevant global weights and their ranking are
shown. The definition and use of local and global weights and how to calculate global weights
are given in section 3.2.7.1 of Chapter 3. For example, the global weight of travel time (C1.1)
is acquired as follows: global weight of C1.1 = local weight of C1.1 × weight of C1 from Table
43; therefore, 0.1047 = 0.2906 × 0.3603.
Table 44: The optimal groups’ weights of government members in each sub-criterion for car-
sharing services.
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall ranking
of sub-criteria
C1
C1.1. Travel time
0.2906
2
0.1047
4
C1.2. Travel distance
0.2036
4
0.0734
6
C1.3. Departure time
0.2097
3
0.0756
5
C1.4. Trip purpose
0.2961
1
0.1067
3
C2
C2.1. Travel cost
0.2894
1
0.1107
2
C2.2. Travel comfort
0.1434
3
0.0548
8
C2.3. Safety
0.1392
5
0.0532
10
C2.4. Service quality
0.1258
6
0.0481
11
C2.5. Environment-friendly system
0.1428
4
0.0546
9
C2.6. User-friendly
0.1594
2
0.0610
7
C3
C3.1. Service availability
0.1553
2
0.0400
12
C3.2. Vehicle availability and
accessibility
0.8447
1
0.2174
1
In this study, the overall rank of the most important sub-criterion determines the starting point
for explaining the sub-criteria. For example, as listed in table 44, the sub-criterion vehicle
availability and accessibility (C3.2) has the best overall ranking (first rank), and this sub-
criterion belongs to the main-criterion availability and accessibility (C3). Hence, the
explanation begins by describing all sub-criteria of category C3 (according to their rank in
category C3). Then, the description of the sub-criteria of category car-sharing characteristics
(C2) is provided (according to their rank in category C2) because travel cost (C2.1) has the
highest overall rank (second rank) among the rest ten sub-criteria and belongs to the main-
criterion C2. Finally, the sub-criteria of the remaining category is explained, which is trip-
related characteristics (C1) (according to their rank in category C1). This table explanation
procedure is also used for related tables in the bike-sharing and scooter-sharing sections.
Table 44 displays that from the perspective of government members, vehicle
availability and accessibility (C3.2) is the most important sub-criterion that individuals
consider in car-sharing usage ( = 0.2174) among the 12 identified sub-criteria. Although
the related main-criterion availability and accessibility (C3) weighs less than the other two
main-criteria trip-related characteristics (C1) and car-sharing characteristics (C2), as shown in
Figure 31, the latter two main-criteria do not have a very high CL compared to C3. Also, only
two sub-criteria were introduced for C3, and the local weight of the sub-criterion C3.2 is much
higher than that of the other sub-criterion (approximately 5.5 times higher). This is not

160
surprising since, in the survey, all government member respondents chose C3.2 as the best sub-
criterion and never selected as the worst sub-criterion. Also, C3.2 and service availability
(C3.1) are the best and the worst sub-criterion out of all 12 sub-criteria, respectively.
Figure 32 displays the credal ranking of sub-criteria belonging to the main-criterion
availability and accessibility (C3). As illustrated in Figure 32, in the eyes of government
members, the sub-criterion vehicle availability and accessibility (C3.2) is absolutely more
important than the sub-criterion service availability (C3.1), with CL equal to 1. As explained
in section 3.2.7.6.4 of Chapter 3, when the threshold value is 50 and CL is above 80, it can be
noted that one criterion is definitely more important than another. This may be because, as
mentioned, all members of the government made the same choice on the best and worst sub-
criterion between these two, although they gave different scores when comparing them.
Figure 32: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (car-sharing services).
Table 44 also demonstrates that among the 12 sub-criteria, travel cost (C2.1) is the second most
important sub-criterion ( = 0.1107). Also, the local weight of the sub-criterion C2.1 is
much higher than the other sub-criteria (about twice) in the category car-sharing characteristics
(C2). Figure 33 indicates the credal ranking of sub-criteria belonging to C2 from the
perspective of government members for car-sharing services. It shows that C2.1 is completely
superior to the other three sub-criteria (CL close to 1). As mentioned in Table 44, the second
most important sub-criterion in this category is user-friendly (C2.6). Furthermore, looking at
Figure 33, it can also be stated that the sub-criterion travel comfort (C2.2) is more important
than the sub-criterion service quality (C2.4) (CL=0.66). However, one cannot be sure of the
sub-criterion C2.2 superiority over the sub-criteria environment-friendly system (C2.5) (0.51)
and safety (C2.3) (0.54). It can also be noted that the sub-criterion C2.5 is more important for
individuals on car-sharing use than the sub-criterion C2.4 (CL=65). However, it cannot be
mentioned that the sub-criterion C2.5 is assuredly perceived as more important than the sub-
criterion C2.3 (CL=0.53). Also, among the six sub-criteria in category C2, C2.4 is the least
important criterion; even C2.3 is ranked higher with a confidence of 0.62.

161
Figure 33: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (car-sharing services).
As listed in Table 44, according to members of the government, among the 12 sub-criteria,
trip purpose (C1.4) is the third most important sub-criterion that plays a role in people's car-
sharing use. Also, C1.4 is the most important sub-criterion in the category trip-related
characteristics (C1). Figure 34 indicates the credal ranking of sub-criteria belonging to the
main-criterion trip-related characteristics (C1). As illustrated in Figure 34, in this category, the
C1.4 is certainly more important than the sub-criteria departure time (C1.3) (CL=0.8) and travel
distance (C1.2) (CL=0.82). Especially since Table 44 indicates that the local weight of the
sub-criterion C1.4 is about 1.5 times higher than these two sub-criteria.
As shown in Figure 34, travel time (C1.1) ranks second in the category trip-related
characteristics (C1), which means it is still more important than departure time (C1.3) and
travel distance (C1.2), with a confidence of about 0.80. It is worth noting that although among
the sub-criteria of the C1 category, people assign the least amount of value to the sub-criterion
C1.2, the sub-criterion C1.3 does not have a high CL compared to the sub-criterion C1.2
(CL=0.53). Also, as presented in Table 44, the local weight of these two sub-criteria is
approximately equal. Therefore, one cannot comment definitively on the superiority of the
sub-criterion C1.3 to the sub-criterion C1.2. Besides, it is important to note that in this
category, the lowest CL is between trip purpose (C1.4) and C1.1, indicating that government

162
members highly value both of these factors when assessing the criteria affecting people's car-
sharing use.
Figure 34: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (car-sharing services).
In summary, to better understand the views of government members on the impact of factors
on people's car-sharing usage, the weight of the three most important sub-criteria, which are
vehicle availability and accessibility (C3.2), travel cost (C2.1), and trip purpose (C1.4),
respectively, and the weight of the least important sub-criterion, which is the service
availability (C3.1), are presented in Figure 35.
Figure 35: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for car-sharing choice).
6.1.1.2 Group weight of operators for car-sharing services
The optimal operators’ group weights of the main-criteria for car-sharing services are listed in
Table 45.
Table 45: Operators’ group weights of the main-criteria for car-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.0963
0.2174 0.1107 0.1067 0.04
0
0.2
0.4
Sub-criteria Weight
Sub-criteria
Government Members' View (Car-sharing)
C3.2 C2.1 C1.4 C3.1

163
Main-criteria
Weights
C2. Car-sharing characteristics
0.4835
C3. Availability and accessibility
0.4203
As presented in Table 45, from the point of view of operators, the most important main-
criterion that individuals could consider in using car-sharing is car-sharing characteristics (C2),
with a weight  = 0.4835. This means that operators believe that people place more value
on main-criterion C2 when using car-sharing rather than on main-criteria trip-related
characteristics (C1) and availability and accessibility (C3). Figure 36 shows the credal ranking
of the main-criteria from the operators' perspective (for car-sharing services) and the assigned
CL.
Figure 36: Credal ranking of main-criteria from operators’ view for car-sharing services.
Figure 36 indicates that the main-criterion car-sharing characteristics (C2) is more important
than the main-criterion availability and accessibility (C3) (CL=0.64), and these two main-
criteria are definitely superior to the main-criterion trip-related characteristics (C1), with CL
equal to 0.99. Also, Table 45 shows that the weight of C2 and C3 is about 4.37 and 5 times
more than that of C1, respectively.
Table 46 gives the optimal group weights of operators for car-sharing services. The
main-criteria followed by the sub-criteria are mentioned. Also, the optimal groups’ local
weights for each sub-criterion and the relevant global weights and their ranking are presented.
Table 46: The optimal groups’ weights of operators in each sub-criterion for car-sharing
services.
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall
ranking of
sub-criteria
C1
C1.1. Travel time
0.2335
2
0.0225
10
C1.2. Travel distance
0.2263
3
0.0218
11
C1.3. Departure time
0.1244
4
0.0120
12
C1.4. Trip purpose
0.4159
1
0.0401
8
C2
C2.1. Travel cost
0.1651
4
0.0798
6
C2.2. Travel comfort
0.1068
5
0.0516
7

164
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall
ranking of
sub-criteria
C2.3. Safety
0.2015
2
0.0974
4
C2.4. Service quality
0.2013
3
0.0973
5
C2.5. Environment-friendly
system
0.0770
6
0.0372
9
C2.6. User-friendly
0.2482
1
0.1200
3
C3
C3.1. Service availability
0.3184
2
0.1338
2
C3.2. Vehicle availability and
accessibility
0.6816
1
0.2865
1
Table 46 displays that from the perspective of operators, vehicle availability and accessibility
(C3.2) is the most important sub-criterion that individuals consider in car-sharing usage (
= 0.2865) among the 12 identified sub-criteria. In addition, it shows that although service
availability (C3.1) is the second most important sub-criteria among all 12 sub-criteria ( =
0.1338), the local weight of sub-criterion C3.2 is twice that of sub-criterion service availability
(C3.1). Besides, Figure 37 displays the credal ranking of the sub-criteria belonging to the main-
criterion availability and accessibility (C3). It shows that the sub-criterion C3.2 is certainly
more important than the sub-criterion C3.1, with Cl equal to 0.91.
Figure 37: Credal ranking of sub-criteria belonging to the main-criterion C3 from operators’
view for car-sharing services.
Table 46 also demonstrates that among the 12 sub-criteria, user-friendly (C2.6) is the third
most important sub-criterion ( = 0.1200). Also, Figure 38 indicates that although sub-
criterion safety (C2.3) is the second most important sub-criterion in the category car-sharing
characteristics (C2), the confidence of 0.5 between sub-criteria C2.3 and service quality (C2.4)
implies that some operators believe that sub-criterion C2.4 plays a more important role. Hence,
the superiority of the sub-criterion C2.3 over the sub-criterion C2.4 is not well established.
Also, among the six sub-criteria in category C2, the environment-friendly system (C2.5) is the
least important criterion; even travel comfort (C2.2) is ranked higher with a confidence of 0.76.

165
Figure 38: Credal ranking of sub-criteria belonging to the main-criterion C2 from operators’
view for car-sharing services.
Figure 39 indicates the credal ranking of sub-criteria belonging to the main-criterion trip-
related characteristics (C1). It shows that sub-criterion trip purpose (C1.4) is certainly more
important than other sub-criteria. Especially since Table 46 indicates that the local weight of
the sub-criterion C1.4 is about 1.8 times higher than that of sub-criterion travel time (C1.1) and
sub-criterion travel distance (C1.2), and it is approximately 3.35 times higher than that of sub-
criterion departure time (C1.3).
Furthermore, Figure 39 indicates that although travel time (C1.1) is the second most
important sub-criterion in the category trip-related characteristics (C1), one cannot comment
definitively on the superiority of the sub-criterion C1.1 to the sub-criterion travel distance
(C1.2). In particular, as listed in Table 46, their local weight is almost equal. In addition, Table

166
46 shows that out of 12 sub-criteria, operators believe that people assign the lowest value to
sub-criterion departure time (C1.3).
Figure 39: Credal ranking of sub-criteria belonging to the main-criterion C1 from operators’
view for car-sharing services.
In summary, to better understand the views of operators on the impact of factors on people's
car-sharing usage, the weight of the three most important sub-criteria, which are vehicle
availability and accessibility (C3.2), service availability (C3.1), and user-friendly (C2.6),
respectively, and the weight of the least important sub-criterion, which is the departure time
(C1.3), are presented in Figure 40.
Figure 40: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of car-sharing operators).
6.1.1.3 Group weight of users for car-sharing services
The optimal users’ group weights of the main-criteria for car-sharing services are listed in
Table 47.
Table 47: Users’ group weights of the main-criteria for car-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.3088
C2. Car-sharing characteristics
0.3089
C3. Availability and accessibility
0.3823
0.2865 0.1338 0.12 0.012
0
0.5
Sub-criteria Weight
Sub-criteria
Operators' View (Car-sharing)
C3.2 C3.1 C2.6 C1.3

167
Table 47 indicates that from the point of view of users, the most important main-criterion that
they could consider in using car-sharing is availability and accessibility (C3), with a weight
 = 0.3823. This implies that from the users' point of view, the most important main-
criterion that can lead them to use car-sharing is C3. Figure 41 shows the credal ranking of the
main-criteria from the users' perspective (for car-sharing services) and the assigned CL.
Figure 41: Credal ranking of main-criteria from users’ view for car-sharing services.
Figure 41 demonstrates that the main-criterion availability and accessibility (C3) is more
important than the main-criterion car-sharing characteristics (C2) and trip-related
characteristics (C1). Also, it shows that one cannot comment on the superiority of the main-
criterion C2 over the main-criterion C1 with Cl equal to 0.5. Especially, Table 47 indicates
that the weights of these two main-criteria are approximately equal.
Table 48 gives the optimal group weights of users of car-sharing services. The main-
criteria followed by the sub-criteria are presented. Also, the optimal groups’ local weights for
each sub-criterion and the relevant global weights and their ranking are listed.
Table 48: The optimal groups’ weights of users in each sub-criterion for car-sharing services.
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall
ranking of
sub-criteria
C1
C1.1. Travel time
0.3760
1
0.1161
3
C1.2. Travel distance
0.2321
2
0.0717
4
C1.3. Departure time
0.1854
4
0.0573
7
C1.4. Trip purpose
0.2065
3
0.0638
6
C2
C2.1. Travel cost
0.2268
1
0.0701
5
C2.2. Travel comfort
0.1418
5
0.0438
11
C2.3. Safety
0.1768
2
0.0546
8
C2.4. Service quality
0.1691
3
0.0522
9
C2.5. Environment-friendly
system
0.1467
4
0.0453
10
C2.6. User-friendly
0.1389
6
0.0429
12
C3
C3.1. Service availability
0.3483
2
0.1332
2
C3.2. Vehicle availability and
accessibility
0.6517
1
0.2491
1

168
Table 48 shows that vehicle availability and accessibility (C3.2) is the most important sub-
criterion that users consider in car-sharing usage ( = 0.2491) among the 12 identified sub-
criteria. Besides, it displays that although service availability (C3.1) is the second most
important sub-criteria among all 12 sub-criteria ( = 0.1332), the local weight of sub-
criterion C3.2 is 1.88 times higher than sub-criterion C3.1. In addition, Figure 42 presents the
credal ranking of the sub-criteria belonging to the main-criterion availability and accessibility
(C3). It shows that the sub-criterion C3.2 is absolutely more important than the sub-criterion
C3.1 with Cl equal to 1.
Figure 42: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for car-sharing services.
Table 48 also establishes that among the 12 sub-criteria, travel time (C1.1) is the third most
important sub-criterion ( = 0.1161). Additionally, Figure 43 indicates that C1.1 is
definitely more important than other sub-criteria in the category trip-related characteristics C1
(CL=1). Similarly, Table 48 shows that in category C1, the local weight of the sub-criterion
C1.1 is almost 1.62 to 2 times higher than that of other sub-criteria. Furthermore, Figure 43
demonstrates that travel distance (C1.2) is the second most important sub-criterion in this
category, which is certainly more important than the trip purpose (C1.4). Both of these sub-
criteria are definitely more important than the sub-criterion departure time (C1.3).

169
Figure 43: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for car-sharing services.
Figure 44 indicates the credal ranking of sub-criteria belonging to the main-criterion car-
sharing characteristics (C2). It reveals that sub-criterion travel cost (C2.1) is certainly more
important than other sub-criteria. Further, although Table 48 indicates that the sub-criterion
user-friendly (C2.6) is the least important sub-criterion in the category C2 ( = 0.0429),
the confidence of 0.57 between the sub-criteria travel comfort (C2.2) and C2.6, as displayed in
Figure 44, implies that some users believe that sub-criterion C2.6 plays a more important role.
Hence, the superiority of the sub-criterion C2.2 over the sub-criterion C2.6 is not well
established.
Figure 44: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of car-sharing services.
In summary, to better understand users' views on the impact of factors on their car-sharing
usage, the weight of the three most important sub-criteria, which are vehicle availability and

170
accessibility (C3.2), service availability (C3.1), and travel time (C1.1), respectively, and the
weight of the least important sub-criterion, which is the user-friendly (C2.6), are presented in
Figure 45.
Figure 45: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of car-sharing).
6.1.1.4 Group weight of non-users for car-sharing services
The optimal non-users’ group weights of the main-criteria for car-sharing services are shown
in Table 49.
Table 49: Non-users’ group weights of the main-criteria for car-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.2465
C2. Car-sharing characteristics
0.3811
C3. Availability and accessibility
0.3724
Table 49 designates that from the non-users perspective, car-sharing characteristics (C2) is the
most important main-criterion that they could consider in using car-sharing, with a weight
 = 0.3811. This implies that from the non-users' standpoint, the most important main-
criterion that can lead them to use car-sharing is C2. Figure 46 displays the credal ranking of
the main-criteria from the non-users' point of view (for car-sharing services) and the assigned
CL.
0.2491
0.1332 0.1161 0.0429
0
0.1
0.2
0.3
Sub-criteria Weight
Sub-criteria
Users' View (Car-sharing)
C3.2 C3.1 C1.1 C2.6

171
Figure 46: Credal ranking of main-criteria from non-users’ view for car-sharing services.
Figure 46 indicates that although the main-criterion car-sharing characteristics (C2) is the most
important main-criterion in this category, one cannot comment on the superiority of the main-
criterion C2 over the main-criterion availability and accessibility (C3) with Cl equal to 0.55.
Table 50 presents the optimal group weights of non-users of car-sharing services. The
main-criteria followed by the sub-criteria are given. Moreover, the optimal groups’ local
weights for each sub-criterion, relevant global weights, and ranking are mentioned.
Table 50: The optimal groups’ weights of non-users in each sub-criterion for car-sharing
services.
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall
ranking of
sub-criteria
C1
C1.1. Travel time
0.3237
1
0.0798
4
C1.2. Travel distance
0.2425
2
0.0598
8
C1.3. Departure time
0.2346
3
0.0578
9
C1.4. Trip purpose
0.1992
4
0.0491
11
C2
C2.1. Travel cost
0.2136
1
0.0814
3
C2.2. Travel comfort
0.1278
6
0.0487
12
C2.3. Safety
0.1729
3
0.0659
6
C2.4. Service quality
0.1730
2
0.0659
5
C2.5. Environment-friendly
system
0.1501
5
0.0572
10
C2.6. User-friendly
0.1626
4
0.0620
7
C3
C3.1. Service availability
0.3528
2
0.1314
2
C3.2. Vehicle availability and
accessibility
0.6472
1
0.2410
1
Table 50 establishes that vehicle availability and accessibility (C3.2) is the most important sub-
criterion that non-users could consider in car-sharing usage ( = 0.2410) among the 12
identified sub-criteria. The sub-criterion C3.2 is 4.95 times more important than travel comfort
(C2.2), the least important sub-criterion. Besides, it indicates that although service availability
(C3.1) is the second most important sub-criteria among all 12 sub-criteria ( = 0.1314), the
local weight of sub-criterion C3.2 is 1.83 times higher than sub-criterion C3.1. Furthermore,
Figure 47 illustrates the credal ranking of the sub-criteria belonging to the main-criterion

172
availability and accessibility (C3). It designates that the sub-criterion C3.2 is definitely more
important than the sub-criterion C3.1 with CL equal to 1.
Figure 47: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-users’
view of car-sharing services.
Table 50 also determines that among the 12 sub-criteria, travel cost (C2.1) is the third most
important sub-criterion ( = 0.0814). In addition, Figure 48 reveals that C2.1 is absolutely
more important than other sub-criteria in the category car-sharing characteristics (C2) (CL=1).
Similarly, Table 50 shows that in category C2, the local weight of the sub-criterion C2.1 is
almost 1.23 to 1.67 times higher than that of other sub-criteria. Additionally, Figure 48
establishes that although service quality (C2.4) is the second most important sub-criterion in
this category, the confidence of 0.50 between the sub-criteria C2.4 and safety (C2.3) implies
that some non-users believe that sub-criterion C2.3 plays a more important role. Hence, the
superiority of the sub-criterion C2.4 over the sub-criterion C2.3 is not well established.
Figure 48: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-users’ view of
car-sharing services.

173
Figure 49 displays the credal ranking of sub-criteria belonging to the main-criterion trip-
related characteristics (C1). It reveals that sub-criterion travel time (C1.1) is surely more
important than other sub-criteria.
Figure 49: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-users’
view of car-sharing services.
In summary, to better understand the standpoint of non-users on the impact of factors on their
car-sharing use, the weight of the three most important sub-criteria, which are vehicle
availability and accessibility (C3.2), service availability (C3.1), and travel cost (C2.1),
respectively, and the weight of the least important sub-criterion, which is the travel comfort
(C2.2), are given in Figure 50.
Figure 50: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of car-sharing).
6.1.1.5 Similarities and differences between the four types of car-sharing stakeholders
In this study, 12 sub-criteria are compared by four different stakeholders to understand their
views on the importance of each sub-criterion that people can consider in using car-sharing.
Some studies in the literature have only worked on the importance of some of these 12 sub-
criteria. However, in this study, all 12 sub-criteria are ranked and compared with each other to
0.241
0.1314 0.0814 0.0487
0
0.1
0.2
0.3
Sub-criteria Weight
Sub-criteria
Non-users View (Car-sharing)
C3.2 C3.1 C2.1 C2.2

174
determine the importance of each sub-criterion compared with other sub-criteria from each
stakeholder's perspective. In addition, most studies have worked on user perspectives only.
However, in this study, these sub-criteria are compared by four groups of stakeholders.
Therefore, the importance of each sub-criterion can be compared from the perspective of four
different stakeholders to distinguish their views on each sub-criterion. This contributes to
knowing the perceptions of different groups of car-sharing stakeholders about the importance
of one main-criterion/sub-criterion (related to the first research question mentioned in Chapter
1).
One of the significant purposes of this study is to determine the gap between the views
of car-sharing stakeholders. In order to designate the difference between the views of
stakeholders, Table 51 indicates the ranking of the main-criteria and sub-criteria corresponding
to each of the stakeholders.
It is important to note that in the literature, sub-criteria service quality (C2.4) and safety
(C2.3), environment-friendly system (C2.5), and user-friendly (C2.6) have not been well
studied. Hence, this study also considers these sub-criteria to figure out the stakeholders' views
on them.
Table 51: Ranking of the main-criteria and sub-criteria corresponding to car-sharing
stakeholders.
Main-
criteria
Ranking of main-criteria corresponding
with car-sharing stakeholders
Sub-criteria
Ranking of sub-criteria corresponding with car-
sharing stakeholders
Government
members
Operators
Users
Non-
users
Government
members
Operators
Users
Non-
users
C1
2
3
3
3
C1.1. Travel time
4
10
3
4
C1.2. Travel
distance
6
11
4
8
C1.3. Departure
time
5
12
7
9
C1.4. Trip
purpose
3
8
6
11
C2
1
1
2
1
C2.1. Travel cost
2
6
5
3
C2.2. Travel
comfort
8
7
11
12
C2.3. Safety
10
4
8
6
C2.4. Service
quality
11
5
9
5
C2.5.
Environment-
friendly system
9
9
10
10
C2.6. User-
friendly
7
3
12
7
C3
3
2
1
2
C3.1. Service
availability
12
2
2
2
C3.2. Vehicle
availability and
accessibility
1
1
1
1
As shown in Table 51, operators and non-users have similar views on the importance of the
main-criteria. There are also considerable similarities in stakeholders' views on the importance
of the sub-criteria. The importance of vehicle availability and accessibility (C3.2) is well-
mentioned in the literature (Brook, 2004; Catalano et al., 2008; Stillwater et al., 2008; Zheng
et al., 2009; Costain et al., 2012; Kim et al., 2017b; Juschten et al., 2017). As indicated in Table
51, all stakeholders believe that C3.2 is the most important sub-criterion among the 12 sub-

175
criteria individuals consider using car-sharing. In addition, some studies have pointed to the
important role of service availability (C3.1) (Millard-Ball, 2005; Shaheen and Rodier, 2005;
Burkhardt and Millard-Ball, 2006; Habib et al., 2012; Kortum and Machemehl, 2012; Kopp et
al., 2015; Wagner et al.,2016; Becker et al., 2017a; Dias et al., 2017; Hu et al., 2018; Namazu
et al., 2018). In this regard, Table 51 shows that in the eyes of users, non-users, and operators,
C3.1 is the second most important sub-criterion. Interestingly, government members and non-
users alike have similar views on the importance of the sub-criterion user-friendly (C2.6). Also,
the environment-friendly system (C2.5) is one of the least important sub-criteria from the point
of view of all stakeholders. In addition, non-users and operators alike emphasize the importance
of service quality (C2.4).
It is also important to pay attention to important differences in the views of
shareholders. As indicated in Table 51, availability and accessibility (C3) is the most important
main-criterion from the users' perspective but the least important from the government
members' view. Unlike all stakeholders who perceive service availability (C3.1) as the second
most important sub-criterion, members of the government consider it the least important sub-
criterion. Remarkably, although the sub-criterion user-friendly (C2.6) is the third most
important sub-criterion from the operators' perspective, users perceive it as the least important
sub-criterion. Besides, compared to government members and users, non-users and operators
place more emphasis on the importance of service quality (C2.4). Moreover, unlike government
members, non-users do not pay attention to the importance of the sub-criterion trip purpose
(C1.4).
Figure 51 and Figure 52 display the weight percentage of the main-criteria and the
global weight percentage of the sub-criteria corresponding with the car-sharing stakeholders,
respectively. This type of result representation has been used in the study of Liu (2016).

176
Figure 51: Importance of main-criteria based on different types of stakeholders.
As shown in Figure 51, main-criterion trip-related characteristics (C1) is 3.21 and 2.56 times
more valuable to users and non-users, respectively, than operators. On the other hand, in
operators' eyes, main-criterion car-sharing characteristics (C2) is 1.57 times and 1.27 times
more important than what is mentioned by users and non-users, respectively.
Furthermore, according to Figures 31, 36, 41, and 46, it can be noted that among the
main-criteria, main-criterion car-sharing characteristics (C2) is the most important main-
criterion for all stakeholders except car-sharing users because they definitely prefer availability
and accessibility (C3) over other main-criteria.
It is also worth noting that from the point of view of operators and non-users, the main-criterion
car-sharing characteristics (C2) is definitely more important than trip-related characteristics
(C1); however, some members of the government prefer C1 to C2.
0.00 10.00 20.00 30.00 40.00 50.00 60.00
C1
C2
C3
Weight Percentage
Main-Criteria
Car-sharing
Non-users Users Operators Government Members

177
Figure 52: Importance of sub-criteria based on different types of stakeholders.
According to Figure 52, it seems that in some cases, the views of operators on the sub-criteria
that people consider in using car-sharing differ from the perspective of users and non-users.
Operators pay 2.8 times and 1.94 times more attention to user-friendly (C2.6) than users and
non-users, respectively. Besides, it is important to note that non-users pay 1.45 times more
attention to C2.6 than users. Also, operators give considerably higher values to sub-criteria
service quality (C2.4) and safety (C2.3) than users and non-users. On the other hand, compared
operators, users, and non-users give 4.8 times and 4.82 times more value to departure time
(C1.3), respectively. Similarly, travel distance (C1.2) and travel time (C1.1) are substantially
more important sub-criteria for users and non-users than operators. Furthermore, Figures 33,
38, 44, and 48 show that in the category car-sharing characteristics (C2), travel cost (C2.1) is
the most important sub-criteria for all stakeholders except operators since they certainly prefer
user-friendly (C2.6) to C2.1. In their view, C2.1 is the fourth most important sub-criterion. It
is worth noting that C1.1 is 1.45 times more important to users than non-users.
Furthermore, according to Figures 34 and 39, it can be pointed out that in the category
trip-related characteristics (C1), for both government members and operators, sub-criteria trip
purpose (C1.4) and travel time (C1.1) are the first and second most important sub-criteria,
respectively. It should be noted that some government members believe that C1.1 is more
important than C1.4, while for operators, C1.4 is definitely more important than C1.1.
Also, Figure 52 shows that the views of government members are also different from
users and non-users. Users and non-users do not value C1.4 as much as government members
deem. Compared to users and non-users, government members give 1.67 times and 2.17 times
more value to C1.4, respectively. In addition, government members pay considerably more
attention to travel cost (C2.1) than users and non-users, respectively. On the other hand, users
0.00 5.00 10.00 15.00 20.00 25.00 30.00
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Car-sharing
Non-users Users Operators Government Members

178
and non-users assign higher importance to service availability (C3.1) than government
members suppose, 3.33 times and 3.29 times higher, respectively. Also, travel distance (C1.2)
is the second most important sub-criteria for users and non-users, especially users, because, as
shown in Figure 43, they certainly prefer C1.2 over C1.4 (the most important sub-criterion
from the perspective of government members in category C1).
6.1.2 Bike-sharing services
In order to determine the priority of each stakeholder of bike-sharing services for the main-
criteria and sub-criteria, the group weight of bike-sharing services stakeholders is analyzed.
6.1.2.1 Group weight of government members for bike-sharing services
Table 52 reveals the optimal government members’ group weights of the main-criteria for bike-
sharing services.
Table 52: Government members’ group weights of the main-criteria for bike-sharing
services.
Main-criteria
Weights
C1. Trip-related characteristics
0.4345
C2. Bike-sharing characteristics
0.2200
C3. Availability and accessibility
0.3455
Table 52 designates that from the government members' perspective, trip-related characteristics
(C1) is the most important main-criterion that individuals could consider in using bike-sharing,
with a weight  = 0.4345. This implies that from the government members' standpoint, the
most important main-criterion that can lead them to use bike-sharing is C1. Figure 53 displays
the credal ranking of the main-criteria from the government members' point of view (for bike-
sharing services) and the assigned CL. The definition of CL is given in section 3.2.7.6.4 of
Chapter 3.
Figure 53: Credal ranking of main-criteria from government members’ view for bike-sharing
services.
Figure 53 indicates that trip-related characteristics (C1) is the most important main-criterion in
this category. It indicates that main-criterion C1 is more important than main-criterion

179
availability and accessibility (C3) (CL=0.64). As mentioned in section 3.2.7.6.4 of Chapter 3,
when the threshold value is 50 and the CL is around 60 to 80, it can be pointed out that one
criterion is more important than the other. Also, when the CL is above 80, it can be noted that
one criterion is definitely more important than another. Hence, C1 is definitely more important
than bike-sharing characteristics (C2) (CL=0.83).
Table 53 presents the optimal group weights of government members for bike-sharing
services. The main-criteria followed by the sub-criteria are given. Moreover, the optimal
groups’ local weights for each sub-criterion, relevant global weights, and ranking are
mentioned.
Table 53: The optimal groups’ weights of government members in each sub-criterion for
bike-sharing services.
Main-
criteria
Sub-criteria
Local weight
per sub-
criterion
Ranking
within
category
Global weight
per sub-
criterion
Overall
ranking of sub-
criteria
C1
C1.1. Travel time
0.4351
1
0.1891
2
C1.2. Travel distance
0.2738
2
0.1190
3
C1.3. Departure time
0.1584
3
0.0688
5
C1.4. Trip purpose
0.1327
4
0.0577
6
C2
C2.1. Travel cost
0.2223
1
0.0489
7
C2.2. Travel comfort
0.1007
6
0.0222
12
C2.3. Safety
0.1784
3
0.0392
9
C2.4. Service quality
0.2147
2
0.0472
8
C2.5. Environment-friendly
system
0.1117
5
0.0246
11
C2.6. User-friendly
0.1722
4
0.0379
10
C3
C3.1. Service availability
0.3161
2
0.1092
4
C3.2. Vehicle availability and
accessibility
0.6839
1
0.2363
1
Table 53 establishes that in the eyes of government members, vehicle availability and
accessibility (C3.2) is the most important sub-criterion that people could consider in bike-
sharing usage ( = 0.2363) among the 12 identified sub-criteria. The sub-criterion C3.2 is
10.64 times more important than travel comfort (C2.2), the least important sub-criterion.
Furthermore, Figure 54 illustrates the credal ranking of the sub-criteria belonging to the main-
criterion availability and accessibility (C3). It designates that the sub-criterion C3.2 is
definitely more important than the sub-criterion service availability (C3.1), with CL equal to
0.95. This is not surprising since, in the main-criterion C3, all government members chose C3.2
as the most important sub-criterion. Besides, it indicates that the C3.1 is the fourth most
important sub-criteria among all 12 sub-criteria ( = 0.1092), the local weight of sub-
criterion C3.2 is 2.16 times higher than sub-criterion C3.1.

180
Figure 54: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (bike-sharing services).
Table 53 also determines that among the 12 sub-criteria, travel time (C1.1) is the second most
important sub-criterion ( = 0.1891). In addition, Figure 55 reveals that C1.1 is absolutely
more important than other sub-criteria in the category trip-related characteristics (C1).
Similarly, Table 53 shows that in category C1, the local weight of the sub-criterion C1.1 is
almost 1.59 to 3.28 times higher than that of other sub-criteria. Additionally, travel distance
(C1.2) is the third most important sub-criterion among the 12 sub-criteria, which is in the main-
criterion trip-related characteristics (C1).
Figure 55: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (bike-sharing services).
As listed in Table 53, according to members of the government, among the 12 sub-criteria, the
travel cost (C2.1) is the seventh most important sub-criterion that plays a role in people's bike-
sharing use. Also, C2.1 is the most important sub-criterion in the category bike-sharing
characteristics (C2). Figure 56 indicates the credal ranking of sub-criteria belonging to the

181
main-criterion C2. Although C2.1 is the most important sub-criterion in category C2, its
superiority over C2.4 is not well-established (CL=0.55). As mentioned in section 3.2.7.6.4 of
Chapter 3, when the threshold value is 50, and the CL is 50 (equal to the threshold value) or
slightly higher (from 50 to less than 60), the superiority of one criterion over another is not
well established. Similarly, a confidence level of 0.55 between C2.3 and C2.6 indicates that
some government members prefer C2.6 to C2.3. Further, the sub-criterion environment-
friendly system (C2.5) is more important than travel comfort (C2.2), and other sub-criteria in
category C2 are definitely more important than C2.2. In particular, Table 53 determines that
among all 12 sub-criteria, government members considered C2.2 as the least important sub-
criterion.
Figure 56: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (bike-sharing services).
In summary, to better understand the standpoint of government members on the impact of
factors on their bike-sharing use, the weight of the three most important sub-criteria, which are
vehicle availability and accessibility (C3.2), travel time (C1.1), and travel distance (C1.2),
respectively, and the weight of the least important sub-criterion, which is the travel comfort
(C2.2), are presented in Figure 57.

182
Figure 57: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for bike-sharing
choice).
6.1.2.2 Group weight of operators for bike-sharing services
The optimal operators’ group weights of the main-criteria for bike-sharing services are
mentioned in Table 54.
Table 54: Operators’ group weights of the main-criteria for bike-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.0967
C2. Bike-sharing characteristics
0.4372
C3. Availability and accessibility
0.4661
Table 54 indicates that operators consider availability and accessibility (C3) with a weight
 = 0.4661 as the most important main-criterion that individuals could consider in using
bike-sharing. Figure 58 displays the credal ranking of the main-criteria from the operators'
point of view (for bike-sharing services) and the assigned CL.
Figure 58: Credal ranking of main-criteria from operators’ view for bike-sharing services.
Figure 58 indicates that although the main-criterion availability and accessibility (C3) is the
most important sub-criterion, some operators believe that the main-criterion bike-sharing
characteristics (C2) is more important (CL=0.56). Also, Table 54 shows that the weight of C3
0.2363 0.1891 0.119 0.0222
0
0.5
Sub-criteria Weight
Sub-criteria
Government Members' View (Bike-sharing)
C3.2 C1.1 C1.2 C2.2

183
and C2 is about 4.82 and 4.52 times more than that of trip-related characteristics (C1),
respectively.
Table 55 provides the optimal group weights of operators for bike-sharing services. The main-
criteria followed by the sub-criteria are mentioned. In addition, the optimal groups’ local
weights for each sub-criterion and the relevant global weights and their ranking are listed.
Table 55: The optimal groups’ weights of operators in each sub-criterion for bike-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.2584
2
0.0250
10
C1.2. Travel distance
0.4307
1
0.0416
9
C1.3. Departure time
0.1898
3
0.0184
11
C1.4. Trip purpose
0.1211
4
0.0117
12
C2
C2.1. Travel cost
0.1141
6
0.0499
8
C2.2. Travel comfort
0.1370
5
0.0599
7
C2.3. Safety
0.1776
2
0.0776
4
C2.4. Service quality
0.1697
3
0.0742
5
C2.5. Environment-friendly
system
0.2366
1
0.1034
3
C2.6. User-friendly
0.1651
4
0.0722
6
C3
C3.1. Service availability
0.5976
1
0.2785
1
C3.2. Vehicle availability
and accessibility
0.4024
2
0.1876
2
Table 55 reveals that operators believe that service availability (C3.1) is the most important
sub-criterion that people consider for using bike-sharing ( = 0.2785) among the 12
identified sub-criteria. Also, the local weight of this sub-criterion is about 1.49 times more
important than that of the sub-criterion vehicle availability and accessibility (C3.2), which is
the second most important sub-criterion among the 12 sub-criteria. Additionally, Figure 59
presents the credal ranking of the sub-criteria belonging to the main-criterion availability and
accessibility (C3). It indicates that sub-criterion C3.1 is more important than sub-criterion C3.2,
with Cl equal to 0.77.

184
Figure 59: Credal ranking of sub-criteria belonging to the main-criterion C3 from operators’
view for bike-sharing services.
Table 55 also establishes that among the 12 sub-criteria, the environment-friendly system
(C2.5) is the third most important sub-criterion ( = 0.1034). Besides, Figure 60 implies
that sub-criterion C2.5 is more important than sub-criterion C2.3 (CL=0.78) and is also
absolutely more important than other sub-criteria in the category bike-sharing characteristics
(C2). Also, Table 55 presents that the local weight of sub-criterion C2.5 is 2.07 times more
important than that of sub-criterion travel cost (C2.1), which is the least important sub-criteria
in category C2. As can be seen in Figure 60, although safety (C2.3) is the second most
important sub-criterion in the category C2, the confidence of 0.55 between this sub-criterion
and sub-criterion service quality (C2.4) implies that some operators believe that sub-criterion
C2.4 plays a more important role. Similarly, some operators consider sub-criterion user-
friendly (C2.6) more important than sub-criterion C2.3 (CL=0.58).

185
Figure 60: Credal ranking of sub-criteria belonging to the main-criterion C2 from operators’
view for bike-sharing services.
Figure 61 indicates the credal ranking of sub-criteria belonging to the main-criterion trip-
related characteristics (C1). It demonstrates that sub-criterion travel distance (C1.2) is
absolutely more important than other sub-criteria in the category trip-related characteristics
(C1). Especially since Table 55 suggests that the local weight of the sub-criterion C1.2 is
approximately 3.56 times higher than that of the sub-criterion trip purpose (C1.4).
Figure 61: Credal ranking of sub-criteria belonging to the main-criterion C1 from operators’
view for bike-sharing services.
In summary, to better understand the perspective of operators on the effect of factors on
individuals’ bike-sharing use, the weight of the three most important sub-criteria, which are
service availability (C3.1), vehicle availability and accessibility (C3.2), and environment-
friendly system (C2.5), respectively, and the weight of the least important sub-criterion, which
is the trip purpose (C1.4), is offered in Figure 62.

186
Figure 62: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of bike-sharing operators).
6.1.2.3 Group weight of users for bike-sharing services
The optimal users’ group weights of the main-criteria for bike-sharing services are stated in
Table 56.
Table 56: Users’ group weights of the main-criteria for bike-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.2245
C2. Bike-sharing characteristics
0.3450
C3. Availability and accessibility
0.4305
Table 56 implies that from the perspective of users, the most important main-criterion that they
could consider in utilizing bike-sharing is availability and accessibility (C3), with a weight
= 0.4305. This indicates that from the users' view, the most important main-criterion that
can motivate them to use bike-sharing is C3. Figure 63 demonstrates the credal ranking of the
main-criteria from the users' point of view (for bike-sharing services) and the assigned CL. It
demonstrates that the main-criterion C3 is certainly the most important main-criterion.
0.2865 0.1338 0.12 0.012
0
0.5
Sub-criteria Weight
Sub-criteria
Operators' View (Bike-sharing)
C3.2 C3.1 C2.6 C1.3

187
Figure 63: Credal ranking of main-criteria from users’ view for bike-sharing services.
Table 57 presents the optimal group weights of users of bike-sharing services. The main-
criteria followed by the sub-criteria are presented. Further, the optimal groups’ local weights
for each sub-criterion and the relevant global weights and their ranking are mentioned.
Table 57: The optimal groups’ weights of users in each sub-criterion for bike-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.3386
1
0.0760
3
C1.2. Travel distance
0.2660
2
0.0597
7
C1.3. Departure time
0.2094
3
0.0470
11
C1.4. Trip purpose
0.1860
4
0.0418
12
C2
C2.1. Travel cost
0.1759
3
0.0607
6
C2.2. Travel comfort
0.1508
5
0.0520
9
C2.3. Safety
0.1791
2
0.0618
5
C2.4. Service quality
0.1831
1
0.0632
4
C2.5. Environment-friendly
system
0.1613
4
0.0556
8
C2.6. User-friendly
0.1499
6
0.0517
10
C3
C3.1. Service availability
0.3877
2
0.1669
2
C3.2. Vehicle availability
and accessibility
0.6123
1
0.2636
1
Table 57 reveals that vehicle availability and accessibility (C3.2) is the most important sub-
criterion that users consider in utilizing bike-sharing ( = 0.2636) among the 12 identified
sub-criteria. Besides, it suggests that although service availability (C3.1) is the second most
important sub-criteria among all 12 sub-criteria ( =0.1669), the local weight of sub-
criterion C3.2 is 1.58 times higher than sub-criterion C3.1. In addition, Figure 64 offers the
credal ranking of the sub-criteria belonging to the main-criterion availability and accessibility
(C3). It shows that the CL is equal to 099.

188
Figure 64: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for bike-sharing services.
Table 57 also determines that among the 12 sub-criteria, travel time (C1.1) is the third most
important sub-criterion ( = 0.0760). Additionally, Figure 65 signifies that C1.1 is
definitely more important than other sub-criteria in the trip-related characteristics (C1)
category. Also, Table 57 indicates that in category C1, the local weight of the sub-criterion
C1.1 is almost 1.27 to 1.82 times higher than that of other sub-criteria. Moreover, Figure 65
establishes that travel distance (C1.2) is the second most important sub-criterion in this
category, which is certainly more important than departure time (C1.3). Both of these sub-
criteria are absolutely more important than the sub-criterion trip purpose (C1.4).
Figure 65: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for bike-sharing services.
Figure 66 designates the credal ranking of sub-criteria belonging to the main-criterion bike-
sharing characteristics (C2). It exposes that although sub-criterion service quality (C2.4) is the
most important sub-criteria in category C2, its superiority over sub-criterion safety (C2.3) is

189
not well established (CL=0.59). Additionally, although Table 57 indicates that the sub-criterion
user-friendly (C2.6) is the least important sub-criterion in the category bike-sharing
characteristics (C2) ( = 0.0517), the confidence of 0.52 between the sub-criteria travel
comfort (C2.2) and C2.6, as shown in Figure 66, suggests that some users believe that sub-
criterion C2.6 plays a more important role. Therefore, the superiority of the sub-criterion C2.2
over the sub-criterion C2.6 is not well established.
Figure 66: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of bike-sharing services.
In summary, to better understand users' standpoint on the impact of factors on their bike-sharing
use, the weight of the three most important sub-criteria, which are vehicle availability and
accessibility (C3.2), service availability (C3.1), and travel time (C1.1), respectively, and the
weight of the least important sub-criterion, which is the user-friendly (C2.6), are displayed in
Figure 67.
Figure 67: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of bike-sharing).
6.1.2.4 Group weight of non-users for bike-sharing services
The optimal non-users’ group weights of the main-criteria for bike-sharing services are
demonstrated in Table 58.
0.2636 0.1669 0.076 0.0418
0
0.5
Sub-criteria Weight
Sub-criteria
Users' View (Bike-sharing)
C3.2 C3.1 C1.1 C1.4

190
Table 58: Non-users’ group weights of the main-criteria for bike-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.3372
C2. Bike-sharing characteristics
0.2798
C3. Availability and accessibility
0.3829
Table 58 indicates that from the non-users view, availability and accessibility (C3) is the most
important main-criterion that they could consider in using bike-sharing, with a weight  =
0.3829. This establishes that from the non-users perspective, the most important main-criterion
that can encourage them to use bike-sharing is C3. Figure 68 illustrates the credal ranking of
the main-criteria from the non-users' view (for bike-sharing services) and the assigned CL.
Figure 68 reveals that main-criterion C3 is more important than main-criterion trip-related
characteristics (C1) (CL=0.78), and both of these main-criteria are certainly more important
than main-criterion bike-sharing characteristics (C2).
Figure 68: Credal ranking of main-criteria from non-users’ view for bike-sharing services.
Table 59 presents the optimal group weights of non-users of bike-sharing services. The main-
criteria followed by the sub-criteria are listed. Additionally, the optimal groups’ local weights
for each sub-criterion, relevant global weights, and ranking are given.
Table 59: The optimal groups’ weights of non-users in each sub-criterion for bike-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.3240
1
0.1093
3
C1.2. Travel distance
0.2608
2
0.0879
4
C1.3. Departure time
0.1946
4
0.0656
6
C1.4. Trip purpose
0.2206
3
0.0744
5
C2
C2.1. Travel cost
0.1797
2
0.0503
8
C2.2. Travel comfort
0.1524
4
0.0426
10
C2.3. Safety
0.2043
1
0.0572
7
C2.4. Service quality
0.1669
3
0.0467
9
C2.5. Environment-friendly
system
0.1520
5
0.0425
11
C2.6. User-friendly
0.1447
6
0.0405
12
C3
C3.1. Service availability
0.3780
1
0.1447
2

191
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C3.2. Vehicle availability
and accessibility
0.6220
2
0.2382
1
Table 59 suggests that the sub-criterion vehicle availability and accessibility (C3.2) is the most
important sub-criterion that non-users could consider in bike-sharing use ( = 0.2382)
among the 12 identified sub-criteria. The sub-criterion C3.2 is 5.88 times more important than
user-friendly (C2.6), the least important sub-criterion. Besides, it indicates that although
service availability (C3.1) is the second most important sub-criteria among all 12 sub-criteria
( = 0.1447), the local weight of sub-criterion C3.2 is almost 1.65 times higher than sub-
criterion C3.1. In this regard, Figure 69 demonstrates the credal ranking of the sub-criteria
belonging to the main-criterion availability and accessibility (C3). It specifies that the sub-
criterion C3.2 is definitely more important than the sub-criterion C3.1 with Cl equal to 1.
Figure 69: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-users’
view of bike-sharing services.
Table 59 also establishes that among the 12 sub-criteria, travel cost (C1.1) is the third most
important sub-criterion ( = 0.1093). Further, Figure 70 presents the credal ranking of sub-
criteria belonging to the main-criterion trip-related characteristics (C1). It shows that sub-
criterion travel time (C1.1) is surely more important than other sub-criteria in category C1.

192
Figure 70: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-users’
view of bike-sharing services.
Figure 71 specifies the credal ranking of sub-criteria belonging to the main-criterion bike-
sharing characteristics (C2). It reveals that C2.3 is definitely more important than other sub-
criteria in category C2. Furthermore, it establishes that although sub-criterion travel comfort
(C2.2) is the fourth most important sub-criterion in this category, the confidence of 0.51
between the sub-criteria C2.2 and environment-friendly system (C2.5) indicates that some non-
users consider that sub-criterion C2.5 plays a more important role. Therefore, the superiority
of the sub-criterion C2.2 over the sub-criterion C2.5 is not well established.
Figure 71: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-users’
view of bike-sharing services.
In summary, to better understand the viewpoint of non-users on the impact of factors on their
bike-sharing usage, the weight of the three most important sub-criteria, which are vehicle
availability and accessibility (C3.2), service availability (C3.1), and travel time (C1.1),
respectively, and the weight of the least important sub-criterion, which is the user-friendly
(C2.6), are given in Figure 72.

193
Figure 72: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of bike-sharing).
6.1.2.5 Similarities and differences between the four types of bike-sharing stakeholders
In this study, 12 sub-criteria are compared by four different stakeholders in order to recognize
their viewpoints on the importance of each sub-criterion that individuals can consider in bike-
sharing usage. In the literature, some research has only focused on the importance of some of
these 12 sub-criteria. However, in this study, all 12 sub-criteria are ranked and compared with
each other to specify the importance of each sub-criterion compared with other sub-criteria
from each stakeholder's standpoint. Additionally, most studies have worked on user
perceptions only. However, in this study, these sub-criteria are compared by four groups of
stakeholders. Hence, the importance of each sub-criterion can be compared from the viewpoint
of four different stakeholders to distinguish their views on each sub-criterion. This contributes
to knowing the perceptions of different groups of bike-sharing stakeholders about the
importance of one main-criterion/sub-criterion (related to the first research question mentioned
in Chapter 1).
One of the main purposes of this study is to clarify the gap between the point of view
of bike-sharing stakeholders. To determine the difference between the perceptions of
stakeholders, Table 60 indicates the ranking of the main-criteria and sub-criteria corresponding
to each of the stakeholders.
It is important to state that in the literature, sub-criteria service quality (C2.4) and safety
(C2.3), environment-friendly system (C2.5), and user-friendly (C2.6) have not been well
researched. Therefore, this study also considers these sub-criteria to determine the stakeholders'
viewpoints on them.
Table 60: Ranking of the main-criteria and sub-criteria corresponding to bike-sharing
stakeholders.
Main-
criteria
Ranking of main-criteria corresponding
with bike-sharing stakeholders
Sub-criteria
Ranking of sub-criteria corresponding with bike-
sharing stakeholders
Government
Members
Operators
Users
Non-
users
Government
Members
Operators
Users
Non-
users
C1
1
3
3
2
C1.1. Travel time
2
10
3
3
C1.2. Travel
distance
3
9
7
4
C1.3. Departure
time
5
11
11
6
C1.4. Trip purpose
6
12
12
5
C2
3
2
2
3
C2.1. Travel cost
7
8
6
8
0.2382 0.1447 0.1093 0.0405
0
0.5
Sub-criteria Weight
Sub-criteria
Non-users' View (Bike-sharing)
C3.2 C3.1 C1.1 C2.6

194
Main-
criteria
Ranking of main-criteria corresponding
with bike-sharing stakeholders
Sub-criteria
Ranking of sub-criteria corresponding with bike-
sharing stakeholders
Government
Members
Operators
Users
Non-
users
Government
Members
Operators
Users
Non-
users
C2.2. Travel
comfort
12
7
9
10
C2.3. Safety
9
4
5
7
C2.4. Service
quality
8
5
4
9
C2.5.
Environment-
friendly system
11
3
8
11
C2.6. User-
friendly
10
6
10
12
C3
2
1
1
1
C3.1. Service
availability
4
1
2
2
C3.2. Vehicle
availability and
accessibility
1
2
1
1
As seen in Table 60, operators and users have similar views on the importance of the main-
criteria. There are also substantial similarities in stakeholders' views on the importance of the
sub-criteria. The importance of vehicle availability and accessibility (C3.2) is stated in the
literature (Froehlich et al., 2008’ Dell'Olio, 2011; Lin and Yang, 2011; Shaheen et al., 2011;
Vogel and Mattfeld, 2011; Bachand-Marleau et al., 2012; Rixey, 2013; Faghih-Imani and
Eluru, 2015; Feng and Li, 2016; Faghih-Imani et al., 2017; Zhang, 2017; Wang and Lindsey,
2019). As specified in Table 60, Government members, users, and non-users believe that C3.2
is the most important sub-criterion among the 12 sub-criteria people consider in bike-sharing
usage.
From the operators' point of view, service availability (C3.1) is the most important sub-
criterion, the second most important sub-criterion in the eyes of users and non-users. In this
regard, some research has pointed to the important role of C3.1 (Vogel et al., 2011; Buck and
Buehler, 2012; Hampshire and Marla, 2012; Kim et al., 2012; Croci and Rossi, 2014; Etienne
and Latifa, 2014; Noland et al., 2016; Wag et al., 2016; El-Assi et al., 2017; Kutela and
Kidando, 2017; Zhang, 2017; Jain et al., 2018; Shen et al., 2018; Duran-Rodas et al., 2019;
Zhao et al., 2019; Ji et al., 2020; Lin et al., 2020). It is noteworthy that government members,
users, and non-users almost alike value the importance of sub-criterion travel time (C1.1). In
the literature, some studies have focused on the importance of this sub-criterion (Krizek et al.,
2005; Garrard et al., 2008; Akar et al., 2013; Kamargianni and Polydoropoulou, 2013; Whalen
et al., 2013’ Dell'Olio et al., 2014; Kamargiani, 2015). Also, in the eyes of both users and
operators, the trip purpose (C1.4) is the least important sub-criterion among the 12 sub-criteria.
There are also important differences in the views of shareholders. In this regard, it
should be noted that despite the belief of bike-sharing operators that people value an
environment-friendly system (C2.5), this sub-criterion is less important for other stakeholders,
especially for government members and non-users. Also, although departure time (C1.3) and
trip purpose (C1.4) are among the least important sub-criteria for users and operators,
government members and non-users emphasize these sub-criteria. In addition, unlike operators
who believe that travel time (C1.1) is less important than most sub-criteria, it is one of the main
sub-criteria from the point of view of other stakeholders.

195
Figure 73 and Figure 74 demonstrate the weight percentage of the main-criteria and the
global weight percentage of the sub-criteria corresponding with the bike-sharing stakeholders,
respectively.
Figure 73: Importance of main-criteria based on different types of stakeholders.
As seen in Figure 73, the importance of main-criterion trip-related characteristics (C1) for
government members is about 4.3 and 1.95 times higher than for operators and users,
respectively. On the other hand, from the operators’ point of view, the importance of main-
criterion bike-sharing characteristics (C2) is 2 and 1.57 times greater for them than for
government members and non-users, respectively.
It is also worth noting that according to Figures 53, 58, 63, and 68, availability and
accessibility of bike-sharing (C3) is the most important main-criterion for all stakeholders
except government members who believe C1 is more important than it. Besides, it should be
noted that some operators believe that the main-criterion C2 is more important than the main-
criterion C3.
010 20 30 40 50
C1
C2
C3
Weight Percentage
Main-Criteria
Bike-sharing
Non-users Users Operators Government Members

196
Figure 74: Importance of sub-criteria based on different types of stakeholders.
In accordance with Figure 44, it appears that in some instances, the point of view of operators
on the sub-criteria that individuals consider in bike-sharing usage differs from the standpoint
of users and non-users. Users and non-users pay roughly 3.57 and 6.36 times more attention to
the trip purpose (C1.4) than operators, respectively. Likewise, users and non-users value
departure time (C1.3), travel distance (C1.2), and travel time (C1.1) considerably more than
operators. On the other hand, the emphasis by operators on the importance of an environment-
friendly system (C2.5) is about 1.86 and 2.43 times higher than that of users and non-users,
respectively. Besides, according to Figures 25, 31, 35, and 40, the main-criterion C1.1 is
absolutely the most important sub-criterion in category C1 from the point of view of all
stakeholders, except for operators who believe that C1.2 is definitely more important than C1.1.
It should also be noted that according to Figures 24, 29, 34, and 39, it should be mentioned that
the sub-criterion vehicle availability and accessibility (C3.2) is definitely more important than
the sub-criterion service availability (C3.1) for all stakeholders except for operators who
believe that the sub-criterion C3.1 is more important than the sub-criterion C3.2.
In some cases, government members also have different views from users and non-
users. Users and non-users pay 2.34 and 1.92 times more attention to travel comfort (C2.2) and
2.26 and 1.73 times more attention to the C2.5 than government members. In contrast,
government members place 2.49 and 1.73 times more value on C1.1 than users and non-users.
Finally, it is interesting to note that according to Figures 56, 60, 66, and 71, none of the
stakeholders have the same priority over the best sub-criterion in category 2.
0 5 10 15 20 25 30
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Bike-sharing
Non-users Users Operators Government Members

197
6.1.3 Scooter-sharing services
In this section, the group weight of scooter-sharing services stakeholders is analyzed in order
to ascertain the priority of each stakeholder of scooter-sharing services for the main-criteria
and sub-criteria.
6.1.3.1 Group weight of government members for scooter-sharing services
Table 61 shows the optimal government members’ group weights of the main-criteria for
scooter-sharing services.
Table 61: Government members’ group weights of the main-criteria for scooter-sharing
services.
Main-criteria
Weights
C1. Trip-related characteristics
0.1334
C2. Scooter-sharing characteristics
0.6236
C3. Availability and accessibility
0.2430
Table 61 indicates that from the government members' perspective, the main-criterion scooter-
sharing characteristics (C2) is the most important main-criterion that people could consider in
scooter-sharing usage, with a weight  = 0.6236. This suggests that from the government
members' viewpoint, the most important main-criterion that can lead them to use scooter-
sharing is C2. Figure 75 displays the credal ranking of the main-criteria from the government
members' standpoint (for scooter-sharing services) and the assigned CL. The definition of CL
is given in section 3.2.7.6.4 of Chapter 3.
Figure 75: Credal ranking of main-criteria from government members’ view for scooter-
sharing services.
As described in section 3.2.7.6.4 of Chapter 3, when the threshold value is 50 and CL is above
80, it can be stated that one criterion is certainly more important than another. Figure 75
indicates that the main-criterion scooter-sharing characteristics (C2) is absolutely the most
important main-criterion in this category.

198
Table 62 offers the optimal group weights of government members for scooter-sharing
services. The main-criteria followed by the sub-criteria are presented. Moreover, the optimal
groups’ local weights for each sub-criterion, relevant global weights, and ranking are revealed.
Table 62: The optimal groups’ weights of government members in each sub-criterion for
scooter-sharing services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.2828
1
0.0377
9
C1.2. Travel distance
0.2408
3
0.0321
11
C1.3. Departure time
0.2122
4
0.0283
12
C1.4. Trip purpose
0.2643
2
0.0353
10
C2
C2.1. Travel cost
0.2029
2
0.1265
3
C2.2. Travel comfort
0.1104
6
0.0688
8
C2.3. Safety
0.2235
1
0.1394
1
C2.4. Service quality
0.1420
5
0.0886
7
C2.5. Environment-friendly
system
0.1506
4
0.0939
6
C2.6. User-friendly
0.1706
3
0.1064
5
C3
C3.1. Service availability
0.4762
2
0.1157
4
C3.2. Vehicle availability
and accessibility
0.5238
1
0.1273
2
Table 62 establishes that safety (C2.3) is the most important sub-criterion that government
members believe people could consider in scooter-sharing usage ( = 0.1394) among the
12 identified sub-criteria. The sub-criterion C2.3 is approximately 4.93 times more important
than the departure time (C1.3), which is the least important sub-criterion ( = 0.0283).
As mentioned in section 3.2.7.6.4 of Chapter 3, when the threshold value is 50, and the
CL is 50 (equal to the threshold value) or slightly higher (from 50 to less than 60), the
superiority of one criterion over another is not well established. Figure 76 indicates the credal
ranking of sub-criteria belonging to the main-criterion scooter-sharing characteristics (C2). It
is important to note that although C2.3 is the most important sub-criterion in the C2 category,
its superiority over C2.1 is not well-established (CL=0.58). Also, when the threshold value is
50, and the CL is around 60 or 80, it can be pointed out that one criterion is more important
than the other. In this category, all sub-criteria are more important than sub-criterion travel
comfort (C2.2), and the local weight of sub-criterion C2.3 is about 2.02 times higher than sub-
criterion C2.2.

199
Figure 76: Credal ranking of sub-criteria belonging to the main-criterion C2 from
government members’ view (scooter-sharing services).
Table 62 also shows that the sub-criterion vehicle availability and accessibility (C3.2) is the
second most important sub-criterion among all 12 sub-criteria ( = 0.1273). However, as
demonstrated in Figure 77, in category availability and accessibility (C3), one cannot comment
on the superiority of C3.2 to C3.1 (CL=0.56), which is the fourth most important sub-criteria
among all 12 sub-criteria ( = 0.1157).
Figure 77: Credal ranking of sub-criteria belonging to the main-criterion C3 from
government members’ view (scooter-sharing services).
Table 62 also reveals that among the 12 sub-criteria, travel time (C1.1) is the ninth most
important sub-criterion ( = 0.0377). Additionally, Figure 78 exposes that although C1.1
is the most important sub-criterion in the trip-related characteristics (C1) category, its
superiority over sub-criterion trip purpose (C1.4) is not well mentioned (CL= 0.55). Likewise,
some government members prefer the sub-criterion travel distance (C1.2) to the sub-criterion
C1.4 (CL= 0.56). Some also consider sub-criterion departure time (C1.3) is more important
than sub-criterion C1.2 (CL= 0.59).

200
Figure 78: Credal ranking of sub-criteria belonging to the main-criterion C1 from
government members’ view (scooter-sharing services).
In summary, to better understand the viewpoint of government members on the influence of
factors on their scooter-sharing usage, the weight of the three most important sub-criteria,
which are safety (C2.3), vehicle availability and accessibility (C3.2), and travel cost (C2.1),
respectively, and the weight of the least important sub-criterion, which is the departure time
(C1.3), are presented in Figure 79.
Figure 79: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of government members for scooter-sharing
choice).
6.1.3.2 Group weight of operators for scooter-sharing services
The optimal operators’ group weights of the main-criteria for scooter-sharing services are listed
in Table 63.
Table 63: Operators’ group weights of the main-criteria for scooter-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.3333
C2. Scooter-sharing characteristics
0.3333
C3. Availability and accessibility
0.3333
0.1394 0.1273 0.1265
0.0283
0
0.1
0.2
Sub-criteria Weight
Sub-criteria
Government Members' View (Scooter-sharing)
C2.3 C3.2 C2.1 C1.3

201
It should be noted that only one scooter-sharing operator’s response (out of the response of
three scooter-sharing operators) about the main criteria has been used. This operator has
considered the importance of these main-criteria equally; the weight of each main-criteria is
0.3333.
Furthermore, it should be mentioned that there is no Figure to display the credal ranking
of the main-criteria from the view of scooter-sharing operators because only one scooter-
sharing operator’s response (out of the response of three scooter-sharing operators) about the
main criteria is used.
Table 64 delivers the optimal group weights of operators for scooter-sharing services. The
main-criteria followed by the sub-criteria are revealed. Additionally, the optimal groups’ local
weights for each sub-criterion and the relevant global weights and their ranking are listed.
Table 64: The optimal groups’ weights of operators in each sub-criterion for scooter-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.1517
3
0.0506
8
C1.2. Travel distance
0.4764
1
0.1588
2
C1.3. Departure time
0.1397
4
0.0466
10
C1.4. Trip purpose
0.2323
2
0.0774
4
C2
C2.1. Travel cost
0.1526
4
0.0509
7
C2.2. Travel comfort
0.1050
6
0.0350
12
C2.3. Safety
0.2616
1
0.0872
3
C2.4. Service quality
0.1415
5
0.0472
4
C2.5. Environment-friendly
system
0.1758
2
0.0586
5
C2.6. User-friendly
0.1636
3
0.0545
6
C3
C3.1. Service availability
0.1274
2
0.0425
11
C3.2. Vehicle availability
and accessibility
0.8726
1
0.2908
1
Table 64 establishes that operators believe that vehicle availability and accessibility (C3.2) is
the most important sub-criterion that individuals consider for scooter-sharing use (=
0.2908) among the 12 identified sub-criteria. Moreover, the local weight of this sub-criterion
is 6.85 times more important than that of the sub-criterion service availability (C3.1), which is
the eleventh most important sub-criterion among the 12 sub-criteria. In this regard, Figure 80
gives the credal ranking of the sub-criteria belonging to the main-criterion availability and
accessibility (C3). It indicates that the sub-criterion C3.2 is certainly more important than the
sub-criterion C3.1 with Cl equal to 1.

202
Figure 80: Credal ranking of sub-criteria belonging to the main-criterion C3 from operators’
view for scooter-sharing services.
Table 64 also exposes that among the 12 sub-criteria, travel distance (C1.2) is the second most
important sub-criterion ( = 0.1588).
Figure 81 signifies the credal ranking of sub-criteria belonging to the main-criterion
trip-related characteristics (C1). It designates that sub-criterion travel distance (C1.2) is
absolutely more important than other sub-criteria in category C1. Especially since Table 64
proposes that the local weight of the sub-criterion C1.2 is nearly 3.41 times higher than that
of the sub-criterion departure time (C1.3). As can be seen in Figure 81, although sub-criterion
C1.3 is the least important sub-criterion in this category, some operators believe that sub-
criterion C1.3 plays a more important role than sub-criterion travel time (C1.1) (CL= 0.56).
Figure 81: Credal ranking of sub-criteria belonging to the main-criterion C1 from operators’
view for scooter-sharing services.
As listed in Table 64, C2.3 is the third most important sub-criteria among all 12 sub-criteria
( = 0.0872). Figure 82 indicates the credal ranking of sub-criteria belonging to the main-

203
criterion scooter-sharing characteristics (C2). Figure 82 implies that sub-criterion safety (C2.3)
is definitely the most important sub-criterion in category C2. Additionally, Table 64 presents
that the local weight of sub-criterion C2.3 is 2.49 times more important than that of sub-
criterion travel comfort (C2.2), which is the least important sub-criteria in category C2. As can
be realized in Figure 82, although the environment-friendly system (C2.5) is the second most
important sub-criterion in category C2, some operators believe that the sub-criterion user-
friendly (C2.6) plays a more prominent role (CL= 0.58). Likewise, although the sub-criterion
C2.6 is the third most important sub-criterion in this category, the sub-criterion travel cost
(C2.1) is more important for some scooter-sharing operators (CL= 0.57). In addition, it is worth
noting that for some scooter-sharing operators, the service quality (C2.4) is more important
than C2.1 (CL=0.58).
Figure 82: Credal ranking of sub-criteria belonging to the main-criterion C2 from operators’
view for scooter-sharing services.
In summary, to better understand the outlook of operators on the effect of factors on
individuals’ scooter-sharing use, the weight of the three most important sub-criteria, which are
vehicle availability and accessibility (C3.2), travel distance (C1.2), and safety (C2.3)
respectively, and the weight of the least important sub-criterion, which is the travel comfort
(C2.2), are shown in Figure 83.

204
Figure 83: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of scooter-sharing operators).
6.1.3.3 Group weight of users for scooter-sharing services
The optimal users’ group weights of the main-criteria for scooter-sharing services are listed in
Table 65.
Table 65: Users’ group weights of the main-criteria for scooter-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.2493
C2. Scooter-sharing characteristics
0.5002
C3. Availability and accessibility
0.2506
Table 65 suggests that from the viewpoint of users, the most important main-criterion that they
could consider in using scooter-sharing is scooter-sharing characteristics (C2), with a weight
 = 0.5002. This implies that from the users' viewpoint, the most important main-criterion
that can lead them to use scooter-sharing is C2, especially since this main-criterion is almost
twice as important as other main-criteria. Figure 84 appears the credal ranking of the main-
criteria from the users' viewpoint (for scooter-sharing services) and the assigned CL.
0.2908
0.1588 0.0872 0.035
0
0.2
0.4
Sub-criteria Weight
Sub-criteria
Operators' View (Scooter-sharing)
C3.2 C1.2 C2.3 C2.2

205
Figure 84: Credal ranking of main-criteria from users’ view for scooter-sharing services.
Figure 84 designates that the main-criterion C2 is definitely the most important main-criterion.
Additionally, it shows that one cannot comment on the superiority of the main-criterion C3
over the main-criterion C1 with Cl equal to 0.51. Especially, Table 65 indicates that the weights
of these two main-criteria are almost equal.
Table 66 determines the optimal group weights of users of scooter-sharing services. The
main-criteria followed by the sub-criteria are presented. As well, the optimal groups’ local
weights for each sub-criterion and the relevant global weights and their ranking are shown.
Table 66: The optimal groups’ weights of users in each sub-criterion for scooter-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.3230
1
0.0805
6
C1.2. Travel distance
0.2658
2
0.0663
9
C1.3. Departure time
0.2075
3
0.0517
11
C1.4. Trip purpose
0.2037
4
0.0508
12
C2
C2.1. Travel cost
0.1872
2
0.0936
3
C2.2. Travel comfort
0.1554
4
0.0777
7
C2.3. Safety
0.2209
1
0.1105
2
C2.4. Service quality
0.1620
3
0.0810
5
C2.5. Environment-friendly
system
0.1447
5
0.0724
8
C2.6. User-friendly
0.1299
6
0.0650
10
C3
C3.1. Service availability
0.3597
2
0.0901
4
C3.2. Vehicle availability
and accessibility
0.6403
1
0.1605
1
Table 66 reveals that vehicle availability and accessibility (C3.2) is the most important sub-
criterion that users consider in scooter-sharing use ( = 0.1605) among the 12 identified
sub-criteria. This sub-criterion is 3.16 times more important than the sub-criterion trip purpose
(C1.4), the least important sub-criterion among all 12 sub-criteria. In addition, Figure 85
introduces the credal ranking of the sub-criteria belonging to the main-criterion availability and
accessibility (C3). It illustrates that the sub-criterion C3.2 is definitely more important than the
C3.1 (CL=1), the fourth most important sub-criterion among all 12 sub-criteria ( =

206
0.0901). In this regard, Table 66 indicates that the local weight of sub-criterion C3.2 is 1.78
times higher than sub-criterion C3.1.
Figure 85: Credal ranking of sub-criteria belonging to the main-criterion C3 from users’
view for scooter-sharing services.
As listed in Table 66, safety (C2.3) is the second most important sub-criterion among all 12
sub-criteria ( = 0.1105). Figure 86 signifies the credal ranking of sub-criteria belonging
to the main-criterion scooter-sharing characteristics (C2). It exposes that sub-criterion safety
(C2.3) is certainly more important than other sub-criteria.
Figure 86: Credal ranking of sub-criteria belonging to the main-criterion C2 from users’
view of scooter-sharing services.
Table 66 also establishes that among the 12 sub-criteria, travel time (C1.1) is the sixth most
important sub-criterion ( = 0.0805). Figure 87 suggests the credal ranking of sub-criteria
belonging to the main-criterion trip-related characteristics (C1). Figure 87 reveals that C1.1 is
absolutely more important than other sub-criteria in the trip-related characteristics (C1)
category. It should be stated that although departure time (C1.3) is the third most important

207
sub-criterion in this category, some users believe that the sub-criterion trip purpose (C1.4) has
more effect than the sub-criterion C1.3 (CL= 0.56).
Figure 87: Credal ranking of sub-criteria belonging to the main-criterion C1 from users’
view for scooter-sharing services.
In summary, to better understand users' sights on the impact of factors on their scooter-sharing
use, the weight of the three most important sub-criteria, which are vehicle availability and
accessibility (C3.2), safety (C2.3), and travel cost (C2.1) respectively, and the weight of the
least important sub-criterion, which is the trip purpose (C1.4), are designated in Figure 88.
Figure 88: The global weight of the least important sub-criterion and the three most
important sub-criteria (from users' perspective of scooter-sharing).
6.1.3.4 Group weight of non-users for scooter-sharing services
The optimal non-users’ group weights of the main-criteria for scooter-sharing services are
displayed in Table 67.
Table 67: Non-users’ group weights of the main-criteria for scooter-sharing services.
Main-criteria
Weights
C1. Trip-related characteristics
0.3399
C2. Scooter-sharing characteristics
0.3372
C3. Availability and accessibility
0.3229
0.1605 0.1105 0.0936 0.0508
0
0.1
0.2
Sub-criteria Weight
Sub-criteria
Users' View (Scooter-sharing)
C3.2 C2.3 C2.1 C1.4

208
Table 67 reveals that from the non-users perspective, trip-related characteristics (C1) is the
most important main-criterion that they could consider in scooter-sharing use, with a weight
 = 0.3399. This implies that from the non-users perspective, the most important main-
criterion that can lead them to use scooter-sharing is C1. Figure 89 exhibits the credal ranking
of the main-criteria from the non-users' view (for scooter-sharing services) and the assigned
CL.
Figure 89: Credal ranking of main-criteria from non-users’ view for scooter-sharing services.
Figure 89 suggests that although C1 is the most important main-criterion in this category, one
cannot comment on the superiority of the main-criterion C1 over the main-criterion C2 with Cl
equal to 0.52.
Table 68 lists the optimal group weights of non-users of scooter-sharing services. The
main-criteria followed by the sub-criteria are provided. Also, the optimal groups’ local weights
for each sub-criterion, relevant global weights, and ranking are stated.
Table 68: The optimal groups’ weights of non-users in each sub-criterion for scooter-sharing
services.
Main-
criteria
Sub-criteria
Local weight per
sub-criterion
Ranking within
category
Global weight per
sub-criterion
Overall ranking of
sub-criteria
C1
C1.1. Travel time
0.3338
1
0.1135
3
C1.2. Travel distance
0.2536
2
0.0862
4
C1.3. Departure time
0.2080
3
0.0707
6
C1.4. Trip purpose
0.2045
4
0.0695
7
C2
C2.1. Travel cost
0.1818
2
0.0613
8
C2.2. Travel comfort
0.1294
6
0.0436
12
C2.3. Safety
0.2100
1
0.0708
5
C2.4. Service quality
0.1717
3
0.0579
9
C2.5. Environment-friendly
system
0.1495
5
0.0504
11
C2.6. User-friendly
0.1575
4
0.0531
10
C3
C3.1. Service availability
0.4061
2
0.1311
2
C3.2. Vehicle availability
and accessibility
0.5939
1
0.1918
1

209
Table 68 determines that vehicle availability and accessibility (C3.2) is the most important sub-
criterion that non-users could consider in scooter-sharing use ( = 0.1918) among the 12
identified sub-criteria. The sub-criterion C3.2 is 4.40 times more important than travel comfort
(C2.2), the least important sub-criterion.
In addition, it indicates that although service availability (C3.1) is the second most
important sub-criteria among all 12 sub-criteria ( = 0.1311), the local weight of sub-
criterion C3.2 is 1.46 times higher than sub-criterion C3.1. Additionally, Figure 90
demonstrates the credal ranking of the sub-criteria belonging to the main-criterion availability
and accessibility (C3). It specifies that the sub-criterion C3.2 is definitely more important than
the sub-criterion C3.1 with CL equal to 1.
Figure 90: Credal ranking of sub-criteria belonging to the main-criterion C3 from non-users’
view of scooter-sharing services.
Figure 91 displays the credal ranking of sub-criteria belonging to the main-criterion trip-related
characteristics (C1). It exposes that sub-criterion travel time (C1.1) is surely more important
than other sub-criteria. Also, although departure time (C1.3) is the third most important sub-
criterion in this category, some non-users believe that sub-criterion trip purpose (C1.4) is more
important than sub-criterion C1.3.

210
Figure 91: Credal ranking of sub-criteria belonging to the main-criterion C1 from non-users’
view of scooter-sharing services.
Table 68 also reveals that among the 12 sub-criteria, safety (C2.3) is the fifth most important
sub-criterion ( = 0.0708). Figure 92 denotes the credal ranking of sub-criteria belonging
to the main-criterion scooter-sharing characteristics (C2). Figure 92 reveals that C2.3 is
certainly more important than other sub-criteria in the C2 category.
Figure 92: Credal ranking of sub-criteria belonging to the main-criterion C2 from non-users’
view of scooter-sharing services.
In summary, to better understand the standpoint of non-users on the impact of factors on their
scooter-sharing use, the weight of the three most important sub-criteria, which are vehicle
availability and accessibility (C3.2), service availability (C3.1), and travel time (C1.1),
respectively, and the weight of the least important sub-criterion, which is the travel comfort
(C2.2), are given in Figure 93.

211
Figure 93: The global weight of the least important sub-criterion and the three most
important sub-criteria (from the perspective of non-users of scooter-sharing).
6.1.3.5 Similarities and differences between the four types of scooter-sharing stakeholders
In this study, 12 sub-criteria are compared by four different stakeholders to realize their point
of view on the importance of each sub-criterion that individuals can consider in scooter-sharing
use. Some studies in the literature have only focused on the importance of some of these 12
sub-criteria. However, in this study, all 12 sub-criteria are ranked and compared with each other
to specify the importance of each sub-criterion compared with other sub-criteria from each
stakeholder's viewpoint. Besides, most research has studied user perspectives only. However,
in this study, these sub-criteria are compared by four groups of stakeholders. This contributes
to knowing the perceptions of different groups of scooter-sharing stakeholders about the
importance of one main-criterion/sub-criterion (related to the first research question mentioned
in Chapter 1).
One of the substantial purposes of this study is to specify the gap between the point of
view of scooter-sharing stakeholders. In order to indicate the difference between the viewpoints
of stakeholders, Table 69 suggests the ranking of the main-criteria and sub-criteria
corresponding to each of the stakeholders.
It is noteworthy that in the literature, sub-criteria service quality (C2.4), environment-
friendly system (C2.5), and user-friendly (C2.6) have not been well examined. Thus, this study
also considers these sub-criteria to realize the stakeholders' points of view on them.
Table 69: Ranking of the main-criteria and sub-criteria corresponding to scooter-sharing
stakeholders.
Main-
criteria
Ranking of main-criteria corresponding
with scooter-sharing stakeholders
Sub-criteria
Ranking of sub-criteria corresponding with
scooter-sharing stakeholders
Government
members
Operators
Users
Non-
users
Government
members
Operators
Users
Non-
users
C1
3
-
3
1
C1.1. Travel time
9
8
6
3
C1.2. Travel
distance
11
2
9
4
C1.3. Departure
time
12
10
11
6
C1.4. Trip purpose
10
4
12
7
C2
1
-
1
2
C2.1. Travel cost
3
7
3
8
C2.2. Travel
comfort
8
12
7
12
C2.3. Safety
1
3
2
5
0.1918 0.1311 0.1135 0.0436
0
0.2
0.4
Sub-criteria Weight
Sub-criteria
Non-users' View (Scooter-sharing)
C3.2 C3.1 C1.1 C2.2

212
Main-
criteria
Ranking of main-criteria corresponding
with scooter-sharing stakeholders
Sub-criteria
Ranking of sub-criteria corresponding with
scooter-sharing stakeholders
Government
members
Operators
Users
Non-
users
Government
members
Operators
Users
Non-
users
C2.4. Service
quality
7
4
5
9
C2.5.
Environment-
friendly system
6
5
8
11
C2.6. User-
friendly
5
6
10
10
C3
2
-
2
3
C3.1. Service
availability
4
11
4
2
C3.2. Vehicle
availability and
accessibility
2
1
1
1
As presented in Table 69, government members and users agree on the importance of the main-
criteria. There are also substantial similarities in stakeholders' beliefs about the importance of
the sub-criteria. The importance of vehicle availability and accessibility (C3.2) is stated in the
literature (Bai and Jiao, 2020; Jiao and Bai, 2020; Popov and Ravi, 2020). As specified in Table
69, C3.2 is the most important sub-criterion among the 12 sub-criteria in the eyes of operators,
users, and non-users. Similarly, government members consider it the second most important
sub-criterion. It is interesting to mention that C2.6 is one of the least important sub-criterion
from the perspective of users and non-users. Also, C2.5 is not an important sub-criterion from
the users' point of view compared to other sub-criteria and is one of the least important sub-
criteria for non-users.
Safety is one of the most important sub-criterion from the perspective of government
members, operators, and especially users. The importance of this sub-criterion is not surprising
since the introduction of the e-scooter-sharing service has created a new risk of injury (Beck et
al., 2020). The importance of this criterion has been explained in the literature (Anderson-Hall
et al., 2019; Berge, 2019; Clewlow, 2019; James et al., 2019; Haworth and Schramm, 2019;
Shaheen and Cohen, 2019; Almannaa et al., 2020; Che et al., 2020; Gössling, 2020; Li et al.,
2020; Riggs and Kawashima, 2020; Ma et al., 2021). However, non-users pay less attention to
this sub-criterion compared to other stakeholders. This could be because they may not have
used the scooter-sharing service. However, they consider it an important sub-criterion (ranking
5), which can even be one reason why non-users do not use the scooter-sharing service.
Interestingly, travel comfort (C2.2) is the least important sub-criterion from the views
of operators and non-users. Besides, government members, operators, and users agree that
departure time (C1.3) is one of the least important sub-criteria. Also, travel distance (C1.2) is
one of the least important sub-criteria from the point of view of government members and
users. However, on the other hand, this sub-criteria is one of the most important sub-criteria
for operators and non-users.
It is also essential to focus on the important differences in the viewpoints of
shareholders. As revealed in Table 69, unlike all stakeholders who perceive service availability
(C3.1) as one of the most important sub-criteria, operators consider it one of the least important
sub-criteria. On the other hand, despite the belief of government members and users that the
trip purpose (C1.4) is one of the least important sub-criterion, operators believe that this sub-

213
criterion plays an important role. In addition, although travel time (C1.1) is one of the most
effective sub-criteria for non-users, other stakeholders do not have similar considerations.
Figure 94 and Figure 95 present the weight percentage of the main-criteria and the
global weight percentage of the sub-criteria corresponding with the scooter-sharing
stakeholders, respectively.
Figure 94: Importance of main-criteria based on different types of stakeholders.
As seen in Figure 94, main-criterion trip-related characteristics (C1) is more than 2.5 times
more valuable to operators and non-users than to government members. On the other hand, in
the government members' eyes, main-criterion scooter-sharing characteristics (C2) is more
than 1.8 times more important than what is mentioned by operators and non-users. Besides, as
displayed in Figures 75, 84, and 89, C2 is definitely the most important main-criterion in the
eyes of government members and users. For non-users, the most important main-criterion is
C1; however, from the perspective of some non-users, C2 is more important than C1.
0 20 40 60 80
C1
C2
C3
Weight Percentage
Main-Criteria
Scooter-sharing
Non-users Users Operators Government Members

214
Figure 95: Importance of sub-criteria based on different types of stakeholders.
Based on Figure 95, it can be pointed out that in some cases, the viewpoints of operators on the
sub-criteria that individuals consider in scooter-sharing use differ from the perspective of users
and non-users. In this regard, it can be stated that operators pay 1.81 and 1.52 times more
attention to vehicle availability and accessibility (C3.2) than users and non-users, respectively.
On the other hand, users and non-users give 2.12 times and 3.08 times higher values to service
availability (C3.1), respectively, than operators. Similarly, users value service quality (C2.4)
and travel comfort (C2.2) considerably more than operators. It should also be noted that the
C2.2 sub-criterion is 1.78 times more important for users than non-users.
Furthermore, as seen in Figures 76, 82, 87, and 92, safety (C2.3) is definitely the most important
sub-criteria from the perspective of operators, users, and non-users. From the point of view of
government members, although C2.3 is the most important sub-criterion, some government
members believe that the sub-criterion travel cost (C2.1) is more important. Also, C2.1 is the
second most important sub-criterion for non-users and definitely for users. However, operators
rank it as the fourth most important sub-criterion.
Moreover, as seen in Figure 81, C1.2 is absolutely the most important sub-criterion,
and C1.1 is the third most important sub-criterion for operators. Even from the perspective of
some operators, sub-criterion C1.3 is more important than sub-criterion C1.1. Also, as seen in
Figure 81, C1.2 is absolutely the most important sub-criterion, and C1.1 is the third most
important sub-criterion for operators. Even from the perspective of some operators, sub-
criterion C1.3 is more important than sub-criterion C1.1. Also, as displayed in Figures 86 and
91, C1.1 is definitely the most important sub-criterion, and C1.2 is certainly the second most
0 5 10 15 20 25 30 35
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Scooter-sharing
Non-users Users Operators Government Members

215
important sub-criterion for users and non-users. Therefore, operators should pay more attention
to sub-criterion C1.1.
The viewpoints of government members are also different from users and non-users.
Users and non-users assign higher importance to sub-criterion travel time (C1.1) than members
of government suppose, 2.14 times and 3.01 times more, respectively. Similarly, users and non-
users pay remarkably more attention to sub-criteria travel distance (C1.2) and departure time
(C1.3) compared to government members. On the other hand, compared to users, government
members pay to sub-criteria travel cost (C2.1), safety (C2.3), environment-friendly system
(C2.5), and user-friendly (C2.6) 2.06 times, 1.97 times, 1.86 times and two times more
attention, respectively. Further, it should be noted that sub-criteria C2.1 and C2.3 are
considerably more valuable to users than non-users.
Furthermore, as displayed in Figures 80, 85, and 90, the sub-criterion C3.2 is absolutely
more important to operators, users, and non-users than sub-criterion C3.1. However, Figure 77
demonstrates that the superiority of C3.2 over C3.1 has not been well established, and some
government members believe sub-criterion C3.1 is more important than C3.2.
6.1.4 Comparing the relative importance of different criteria among the
three types of shared mobility services
In this section, the differences in the importance of one main-criterion/sub-criterion across the
three types of shared mobility services, including car-sharing, bike-sharing, and scooter-
sharing, are examined for each specific type of stakeholder (government members, operators,
users, and non-users). This contributes to understanding how one main-criterion/sub-criterion
can be of different importance across different shared mobility services (related to the second
research question mentioned in Chapter 1). Hence, these differences in the importance of one
main-criterion/sub-criterion across the three types of shared mobility services should be
considered from each stakeholder's standpoint.
6.1.4.1 From the perspective of government members
The importance (weight percentage) of the main-criteria and sub-criterion based on different
shared mobility services from the views of government members is displayed in Figure 96 and
Figure 97, respectively. It should be noted that the government members who responded to the
car-sharing survey may be different from the government members who responded to the bike-
sharing or scooter-sharing survey.

216
Figure 96: Importance of main-criteria based on different shared mobility services from the
government members' views.
As displayed in Figure 96, government members see the main-criterion scooter-sharing
characteristics (C2) in scooter-sharing (respondents of the scooter-sharing survey) service as
the most important, and the main-criterion trip-related characteristics (C1) in scooter-sharing
received (by the respondents of the scooter-sharing survey) the lowest importance. Besides, the
importance of the main-criterion C2 in the scooter-sharing service (received by the respondents
of the scooter-sharing survey) is approximately 2.83 times greater than its importance in the
bike-sharing service (received by the respondents of the bike-sharing survey). On the other
hand, the importance of the main-criterion C1 in bike-sharing (received by the respondents of
the bike-sharing survey) is almost 3.26 times more than its importance in scooter-sharing
(received by the respondents of the scooter-sharing survey).
0 20 40 60 80
C1
C2
C3
Weight Percentage
Main-Criteria
Government Members
Scooter-sharing Bike-sharing Car-sharing

217
Figure 97: Importance of sub-criteria based on different shared mobility services from the
government members' views.
Figure 97 displays that from the government members’ view, among the type of shared
mobility services, the importance of the sub-criterion vehicle availability and accessibility
(C3.2) in bike-sharing (received by the respondents of the bike-sharing survey) is the highest
among all sub-criteria. It is 10.64 times more important than the sub-criterion travel comfort
(C2.2) in bike-sharing (the least important sub-criterion) (received by the respondents of the
bike-sharing survey). When it comes to sub-criterion travel time (C1.1), its importance in bike-
sharing (received by the respondents of the bike-sharing survey) is 5.02 times more than that
of in scooter-sharing (received by the respondents of the scooter-sharing survey). Similarly,
sub-criterion travel distance (C1.2) in bike-sharing (received by the respondents of the bike-
sharing survey) is 3.71 times more important than the one in scooter-sharing (received by the
respondents of the scooter-sharing survey). On the other hand, the importance of sub-criteria
C2.2, safety (C2.3), and environment-friendly system (C2.5) in scooter-sharing (received by
the respondents of the scooter-sharing survey) are 3.10, 3.56, and 3.82 times more than their
importance in bike-sharing (received by the respondents of the bike-sharing survey),
respectively. It is also worth noting that the importance of departure time (C1.4) in car-sharing
(received by the respondents of the car-sharing survey) is 3.02 times greater than its importance
in scooter-sharing (received by the respondents of the scooter-sharing survey).
6.1.4.2 From the perspective of operators
The importance (weight percentage) of the main-criteria and sub-criterion based on different
shared mobility services from the views of different operators is displayed in Figure 98 and
0 5 10 15 20 25
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Government Members
Scooter-sharing Bike-sharing Car-sharing

218
Figure 99, respectively. It is important to note that the operators of each shared mobility service
are different from the operators of other shared mobility services.
Regarding figure 98, it is apparent that the importance of the main-criterion C1 in scooter-
sharing (received by scooter-sharing operators) is about 3.45 times more than its importance in
bike-sharing and car-sharing (received by bike-sharing and car-sharing operators).
Figure 98: Importance of main-criteria based on different shared mobility services from the
operators' views.
Concerning Figure 99, it is noticeable that the importance of the sub-criterion vehicle
availability and accessibility (C3.2) in scooter-sharing (received by scooter-sharing operators)
is the most among all sub-criteria and types of shared mobility services. It is roughly 24.85
times more than the importance of the sub-criterion trip purpose (C1.4) in bike-sharing
(received by bike-sharing operators), which is of the least importance. Also, the importance of
sub-criterion C1.4 in scooter-sharing (received by scooter-sharing operators) is about 6.62
times more than its importance in bike-sharing (received by bike-sharing operators). On the
other hand, the sub-criterion service availability (C3.1.) in bike-sharing received (by bike-
sharing operators) 6.55 times more attention than C3.1 in scooter-sharing (received by scooter-
sharing operators). In addition, it should be noted that the sub-criteria travel distance (C1.2)
and departure time (C1.3) in scooter-sharing (in the eyes of scooter-sharing operators) are about
7.28 and 3.88 times more important than in car-sharing (in the eyes of car-sharing operators),
respectively.
0 10 20 30 40 50 60
C1
C2
C3
Weight Percentage
Main-Criteria
Operators
Scooter-sharing Bike-sharing Car-sharing

219
Figure 99: Importance of sub-criteria based on different shared mobility services from the
operators' views.
6.1.4.3 From the perspective of users
The importance (weight percentage) of the main-criteria and sub-criterion according to
different shared mobility services from the views of their users is displayed in Figure 100 and
Figure 101, respectively. It is important to note that users of each shared mobility service can
be different from users of other shared mobility services.
In relation to Figure 100, it is noteworthy that among the types of shared mobility
service and all the main-criteria, the importance of the main-criterion scooter-sharing
characteristics (C2) in scooter-sharing (from the scooter-sharing users' view) (the most
important sub-criterion), is 2.24 times more than main-criterion trip-related characteristics (C1)
in bike-sharing (from the bike-sharing users' perspective) (the least important sub-criterion).
0 5 10 15 20 25 30 35
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Operators
Scooter-sharing Bike-sharing Car-sharing

220
Figure 100: Importance of main-criteria based on different shared mobility services from the
users' views.
Given Figure 101, from the perspective of users, it is clear that, in comparison with other sub-
criteria in all shared mobility services, the sub-criterion vehicle availability and accessibility
(C3.2) always receive the greatest value, regardless of which shared mobility service.
Particularly, it attracts the greatest attention in the bike-sharing service (from the bike-sharing
users' perspective). Besides, the sub-criterion C3.2 in bike-sharing received (by bike-sharing
users) received 6.31 times more attention than the sub-criterion trip purpose (C1.4) in bike-
sharing (by bike-sharing users), which is the least important sub-criterion. Moreover, the
importance of the sub-criterion safety (C2.3) in scooter-sharing (from the scooter-sharing users'
view) is 1.78 and 2 times higher than that of sub-criterion C2.3 in car-sharing and bike-sharing
(from the car-sharing and bike-sharing users' views). This may indicate that scooter-sharing
users are more concerned about the safety issues of scooter-sharing services.
0 10 20 30 40 50 60
C1
C2
C3
Weight Percentage
Main-Criteria
Users
Scooter-sharing Bike-sharing Car-sharing

221
Figure 101: Importance of sub-criteria based on different shared mobility services from the
users' views.
6.1.4.4 From the perspective of non-users
The importance (weight percentage) of the main-criteria and sub-criterion according to
different shared mobility services from the views of their non-users is displayed in Figure 102
and Figure 103, respectively. It is important to note that non-users of each shared mobility
service can be different from non-users of other shared mobility services.
According to Figure 102, it is remarkable that among the types of shared mobility
services and all the main-criteria, the main-criterion availability and accessibility (C3) in bike-
sharing (from the non-users of the bike-sharing service view) is the most important. It is
slightly more important than the car-sharing characteristics (C2) in car-sharing (from the non-
users of car-sharing services view).
0 5 10 15 20 25 30
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Users
Scooter-sharing Bike-sharing Car-sharing

222
Figure 102: Importance of main-criteria based on different shared mobility services from the
non-users' views.
According to Figure 103, it is evident that, from the non-users' point of view, compared to other
sub-criteria in all shared mobility services, the sub-criterion vehicle availability and
accessibility (C3.2) always receives the highest importance no matter in which shared mobility
service. In particular, it received (by the car-sharing non-users view) the greatest attention in
the car-sharing service. The sub-criterion C3.2 in car-sharing (from the car-sharing non-users
view) is 5.95 times more important than sub-criterion user-friendly (C2.6) in bike-sharing
(from the bike-sharing non-users perspective).
0 10 20 30 40 50
C1
C2
C3
Weight Percentage
Main-Criteria
Non-users
Scooter-sharing Bike-sharing Car-sharing

223
Figure 103: Importance of sub-criteria based on different shared mobility services from the
non-users' views.
6.1.4.5 Summary of comparison between the views of the four stakeholders related to
different services
This part summarizes the comparisons of the views of the four stakeholders on the main-
criteria/sub-criteria related to different services. Concerning car-sharing services, it should be
noticed that Figure 96 indicates that from the standpoints of government members (respondents
of the car-sharing survey), users, and non-users of car-sharing services, there is no considerable
difference in the importance of the main-criteria. However, as indicated in Figure 98, car-
sharing operators believe that the main-criteria car-sharing characteristics (C2) and availability
and accessibility (C3) receive 5.02 and 4.36 times more attention than the main-criterion trip-
related characteristics (C1). Also, as shown in figure 97, government members (respondents of
the car-sharing survey) give at least two times more attention to the sub-criteria trip purpose
(C1.4), travel cost (C2.1), and vehicle availability and accessibility (C3.2) than the sub-criteria
safety (C2.3), service quality (C2.4), and service availability (C.3.1). In addition, Figure 99
indicates that car-sharing operators believe that sub-criteria user-friendly (C2.6), service
availability (C3.1), and vehicle availability and accessibility (C3.2) are at least two times more
important than the sub-criteria travel time (C1.1), travel distance (C1.2), departure time (C1.3),
trip purpose (C1.4), travel comfort (C2.2), and environment-friendly system (C2.5). In
addition, Figure 101 illustrates that from the car-sharing users' view, sub-criteria travel time
(C1.1), service availability (C3.1), and vehicle availability and accessibility (C3.2) are at least
twice as important as sub-criteria departure time (C1.3), travel comfort (C2.2), safety (C2.3),
service quality (C2..4), environment-friendly system (C2.5), and user-friendly (C2.6).
Additionally, Figure 103 indicates that from the standpoint of non-users of car-sharing, the sub-
0 5 10 15 20 25 30
C1.1
C1.2
C1.3
C1.4
C2.1
C2.2
C2.3
C2.4
C2.5
C2.6
C3.1
C3.2
Global Weight Percentage
Sub-criteria
Non-users
Scooter-sharing Bike-sharing Car-sharing

224
criteria service availability (C3.1) and vehicle availability and accessibility (C3.2) receive at
least two times more attention than sub-criteria travel distance (C1.2), departure time (C1.3),
trip purpose (C1.4), travel comfort (C2.2), environment-friendly system (C2.5), and user-
friendly (C2.6).
Furthermore, for bike-sharing services, it is important to note that in the eyes of
government members (respondents of the bike-sharing survey), the importance of main-
criterion trip-related characteristics (C1) is about 1.98 times greater than that of main-criterion
car-sharing characteristics (C2), as seen in Figure 96. However, Figure 98 displays that from
the point of view of bike-sharing operators, the importance of main-criterion availability and
accessibility (C3) and car-sharing characteristics (C2) is 4.82 and 4.52 times more than that of
main-criterion trip-related characteristics (C1). Also, Figure 100 suggests that for bike-sharing
users, the importance of main-criterion availability and accessibility (C3) is 1.92 times more
than main-criterion trip-related characteristics (C1). On the other hand, Figure 102 delivers that
for non-users of bike-sharing services, there is not much difference in the importance of the
main-criteria. Also, as seen in Figure 97, government members pay at least two times more
attention to sub-criteria travel time (C1.1), travel distance (C1.2), service availability (C3.1),
and vehicle availability and accessibility (C3.2) than to the sub-criteria travel cost (C2.1), travel
comfort (C2.2), safety (C2.3), service quality (C2.4), environment-friendly system (C2.5), and
user-friendly (C2.6). Also, as revealed in figure 99, in the eyes of bike-sharing operators, sub-
criteria environment-friendly system (C2.5), service availability (C3.1), and vehicle
availability and accessibility (C3.2) are at least two times more important than the sub-criteria
travel time (C1.1), travel distance (C1.2), departure time (C1.3), trip purpose (C1.4), and travel
cost (C2.1). Also, as seen in Figure 101, bike-sharing users believe that sub-criteria service
availability (C3.1) and vehicle availability and accessibility (C3.2) are at least two times more
important than other sub-criteria. Besides, as shown in Figure 103, bike-sharing non-users pay
at least two times more attention to the sub-criteria travel time (C1.1), service availability
(C3.1), and vehicle availability and accessibility (C3.2) than the sub-criteria travel cost (C2.1),
travel comfort (C2.2), service quality (C2.4), environment-friendly system (C2.5), and user-
friendly (C2.6).
Moreover, according to the scooter-sharing services, Figure 96 indicates that
government members (respondents of the scooter-sharing survey) pay 4.67 and 2.57 times
more attention to the main-criterion car-sharing characteristics (C2) than the main-criteria trip-
related characteristics (C1) and availability and accessibility (C3), respectively. Similarly, as
shown in Figure 100, scooter-sharing users give almost twice as much importance to the main-
criterion car-sharing characteristics (C2) as the main-criteria trip-related characteristics (C1)
and availability and accessibility (C3). On the other hand, as demonstrated in Figures 98 and
102, from the point of view of both operators and non-users of scooter-sharing services, the
importance of the main-criteria is similar. Besides, as displayed in Figure 97, government
members place twice more value on sub-criteria travel cost (C2.1), safety (C2.3), user-friendly
(C2.6), service availability (C3.1), and vehicle availability and accessibility (C3.2) than on sub-
criteria travel time (C1.1), travel distance (C1.2), departure time (C1.3), and trip purpose
(C1.4). Further, Figure 99 suggests that from the scooter-sharing operators' perspective, sub-
criteria travel distance (C1.2) and vehicle availability and accessibility (C3.2) are at least two

225
times more important than the sub-criteria travel time (C1.1), departure time (C1.3), trip
purpose (C1.4), travel cost (C2.1), travel comfort (C2.2), service quality (C2.4), environment-
friendly system (C2.5), user-friendly (C2.6), and service availability (C3.1). Moreover, Figure
101 reveals that in the eyes of scooter-sharing users, the sub-criteria safety (C2.3) and vehicle
availability and accessibility (C3.2) receive at least twice more attention than sub-criteria
departure time (C1.3) and trip purpose (C1.4). In addition, Figure 103 illustrates that from the
non-users of scooter-sharing perspective, sub-criteria travel time (C1.1), service availability
(C3.1), and vehicle availability and accessibility (C3.2) have at least two times more value than
sub-criteria travel comfort (C2.2), environment-friendly system (C2.5), and user-friendly
(C2.6).
Finally, from the point of view of each stakeholder (regardless of the type of shared
mobility service), it is worth summarizing which sub-criteria are at least twice as important as
the other sub-criteria (if any). Figure 99 indicates that operators of all shared mobility services
believe that the sub-criterion vehicle availability and accessibility (C3.2) is at least two times
more important than the sub-criteria travel time (C1.1), departure time (C1.3), and trip purpose
(C1.4). In addition, Figure 101 illustrates that from the users' view of all shared mobility
services, the sub-criteria vehicle availability and accessibility (C3.2) is at least twice as
important as the sub-criteria departure time (C1.3). Therefore, operators and users of all shared
mobility services agree that the sub-criterion vehicle availability and accessibility (C3.2) is at
least twice as important as the sub-criterion departure time (C1.3). Additionally, as shown in
Figure 103 indicates that from the standpoint of non-users of all shared mobility services, the
sub-criteria service availability (C3.1) and vehicle availability and accessibility (C3.2) receive
at least two times more attention than sub-criteria travel comfort (C2.2), environment-friendly
system (C2.5), and user-friendly (C2.6).
6.2 Results of the Analysis for Shared Mobility Services (as a whole,
not for a specific shared mobility service)
This section aims to determine the perception of four important shared mobility services (as a
whole, not for a specific shared mobility service) by different stakeholders (government
members, shared mobility operators, shared mobility services users, and non-users). In
particular, it is important to determine which factors are the most important criteria that drive
government members' choice in deciding on a new shared mobility system to be set up in Turin,
Italy. Also, it is important to know which factors are the most important criteria that can drive
operators' choices in planning to run their shared mobility system in a city. Further, which
factors are the most important criteria that users and non-users could consider when selecting
shared mobility to make a trip should be understood. It is important to note that in this section,
only the criteria that can be quantified in this study are considered.
Therefore, this section helps to know how different stakeholders score the importance
of the comparison factors related to themselves (the third research question mentioned in
Chapter 1). Also, it contributes to understanding which shared mobility system is most
appropriate to implement according to users' and non-users’ perceptions (fourth research

226
question mentioned in Chapter 1). Besides, the same criteria are compared by both users and
non-users. Therefore, the importance of these criteria can be compared from the standpoint of
both users and non-users to distinguish their perspectives on each criterion. This help to know
the perceptions of different groups about the importance of one criterion (related to the first
research question mentioned in Chapter 1). In addition, scenarios are presented from the views
of users and non-users groups to determine how to increase the use of bike-sharing and scooter-
sharing services compared to car-sharing services from users’ and non-users’ perspectives
(related to the fifth research question mentioned in Chapter 1).
In this section, four different parts are presented to analyze the views of each
stakeholder (government members, operators, users, and non-users) about shared mobility
services (as a whole, not for a specific shared mobility service). Also, subsection 6.2.5 explains
the similarities and differences between the four types of stakeholders of shared mobility
services (as a whole, not for a specific shared mobility service). Perception analysis (mentioned
in section 3 of Chapter 3) (for users' and non-users’ perspectives) and sensitivity analysis (for
users' and non-users' views) on the former results are finally provided in subsections 6.2.6 and
6.2.7, respectively.
6.2.1 Group weight of government members (shared mobility services as a
whole, not for a specific shared mobility service)
The optimal government members’ group weights of criteria for shared mobility services are
listed in Table 70.
Table 70: Government members’ group weights of criteria for shared mobility services.
Criteria
Weights
Ranking of criteria
Cg1. Average number of trips per vehicle per day
0.1488
4
Cg2. Greenhouse gases (GHGs)
0.1528
3
Cg3. Parking issues
0.1372
5
Cg4. Emission of pollutants
0.1996
2
Cg5. Integration of the shared mobility service with public transport
0.2505
1
Cg6. Vehicle fee
0.1111
6
As stated in Table 70, the integration of the shared mobility service with public transport (Cg5)
is the most important criterion among the six identified criteria that drives government
members' choice in deciding on a new shared mobility system to be set up in Turin, Italy, with
a weight  = 0.2505. It can be explained that criterion Cg5 represents the complementarity
of a shared vehicle for public transport, the integration of which can increase urban mobility.
The second most important criterion is the emission of pollutants (Cg4) which is the amount
of greenhouse gas emissions by a shared mobility system. The importance of this criterion is
about 80% of the importance of the most important criterion, which is a sign of the importance
of this criterion for the members of the government. Similarly, the greenhouse gas (GHGs)
(Cg2) criterion is remarkable for government members, the third most important criterion. It is
not surprising since this criterion shows the pollutants a shared vehicle emits, and governmental
members value sustainability and strive for fewer emissions.

227
The fourth most important criterion is the average number of trips per vehicle per day
(Cg1), which provides insight into the efficiency of the vehicle that shows service efficiency.
The criterion parking issues (Cg3) is illegal parking of shared vehicles, such as parking in
inappropriate places, which is the fifth most important criterion in the eyes of government
members. Finally, the least important criterion is the vehicle fee (Cg6), which is the fee that a
shared mobility operator may pay to the municipality. For example, car-sharing operators may
pay a fee to the municipality that allows their shared cars to go to city centers or places where
traffic is restricted. As presented in Table 70, criterion Cg5 is 2.25 times more important than
criterion Cg6.
Figure 104 reveals the credal ranking of the criteria from the government members'
view (for shared mobility services) and the assigned CL. The definition of CL is given in
section 3.2.7.6.4 of Chapter 3.
Figure 104: Credal ranking of criteria from government members’ view for shared mobility
services.
Figure 104 implies that the integration of the shared mobility service with public transport
(Cg5) is definitely the most important criterion. As mentioned in section 3.2.7.6.4 of Chapter
3, when the threshold value is 50, and the CL is above 80, it can be noted that one criterion is
definitely more important than another. Besides, as mentioned in section 3.2.7.6.4 of Chapter
3, when the threshold value is 50 and the CL is around 60 to 80, it can be pointed out that one
criterion is more important than the other. Hence, it can be mentioned that Cg1 is more
important than Cg3 (CL=0.65). Similarly, Cg2 is more important than Cg3 (CL=0.70).
However, a confidence level of 0.55 between Cg2 and Cg1 indicates that some government
members prefer Cg1 to Cg2. This is because when the threshold value is 50, and the CL is 50
(equal to the threshold value) or slightly higher (from 50 to less than 60), the superiority of one
criterion over another is not well established.
In summary, to better understand the standpoint of government members on the impact
of factors on their decision to set up a new shared mobility service in Turin, Italy, the weight
of the three most important criteria, which are the integration of the shared mobility service

228
with public transport (Cg5), emission of pollutants (Cg4), and greenhouse gases (GHGs) (Cg2),
respectively, and the weight of the least important criterion, which is the vehicle fee (Cg6), are
offered in Figure 105.
Figure 105: The weight of the least important criterion and the three most important criteria
(from the perspective of government members for shared mobility choice).
6.2.2 Group weight of operators of shared mobility services (as a whole, not
for a specific shared mobility service)
The optimal operators’ group weights of the criteria for shared mobility services are mentioned
in Table 71.
Table 71: Operators’ group weights of the criteria for shared mobility services.
Criteria
Weights
Ranking of criteria
Co1. Vehicle utilization rate
0.2916
1
Co2. Usage fees
0.2756
2
Co3. Average number of trips per vehicle per day
0.2606
3
Co4. Operational speed
0.0890
4
Co5. The life span of the vehicle
0.0832
5
Table 71 indicates that vehicle utilization rate (Co1) with a weight  = 0.2916 is the most
important criterion among the five identified criteria that can drive operators' choice in planning
to run their shared mobility system in a city. Average usage rate (%) means the total usage time
of shared vehicles per day divided by their potential usage time per day in 24 hours. This is not
surprising because the vehicle utilization rate is related to the efficiency of their services. Also,
the usage fee (Co2) is the operators' second most important criterion. This could be because it
affects their earnings. In this analysis, operators were supposed to be free to set the price of
their services. Besides, the average number of trips per vehicle per day (Co3) is the third most
important criterion for operators that may be because it gives insight into the vehicle's
efficiency showing the service's efficiency. One of the criteria that received less importance
from operators (the fourth most important criterion) is the operational speed (Co4), which is
the average velocity that a shared mobility system overpasses. Also, the criterion life span of
the vehicle (Co5) is the least important criterion for operators. The system lifespan can be
measured in terms of years and is indicated by the lifespan of vehicles. Moreover, as presented
in Table 71, the criterion Co1 is 3.5 times more important than the criterion Co5.
0.2505 0.1996 0.1528 0.1111
0
0.5
Criteria Weight
Criteria
Government Members' View (Shared Mobility)
C5 C4 C2 C6

229
Figure 106 displays the credal ranking of the criteria from the operators’ point of view
(for shared mobility services) and the assigned CL.
Figure 106: Credal ranking of criteria from operators’ view for shared mobility services.
Figure 106 also suggests that the criterion vehicle utilization rate (Co1) is more important than
the criteria usage fee (Co2) and the average number of trips per vehicle per day (Co3) and is
certainly more important than the criteria operational speed (Co4) and the life span of the
vehicle (Co5). As mentioned in section 3.2.7.6.4 of Chapter 3, when the threshold value is 50
and the CL is around 60 or 80, it can be pointed out that one criterion is more important than
the other. Also, when the threshold value is 50 and the CL is above 80, it can be noted that one
criterion is certainly more important than another.
In summary, to better understand the standpoint of operators on the effect of factors
that can drive operator’s choice in planning to run their shared mobility system in a city, the
weight of the three most important criteria, which are vehicle utilization rate (Co1), usage fees
(Co2), and the average number of trips per vehicle per day (Co3), respectively, and the weight
of the least important criterion, which is the life span of the vehicle (Co5), are presented in
Figure 107.

230
Figure 107: The weight of the least important criterion and the three most important criteria
(from the perspective of shared mobility operators).
6.2.3 Group weight of users of shared mobility services (as a whole, not for a
specific shared mobility service)
The optimal users’ group weights of the criteria for shared mobility services are listed in Table
72.
Table 72: Users’ group weights of the criteria for shared mobility services.
Criteria
Weights
Ranking of criteria
Cp1. Traveler safety
0.1781
1
Cp2. Operational speed
0.1229
5
Cp3. Accessibility
0.1385
3
Cp4. User-friendliness
0.1171
6
Cp5. Image
0.0694
8
Cp6. Comfort
0.1260
4
Cp7. Cost
0.1437
2
Cp8. Possibility of carrying items
0.1042
7
Table 72 indicates that traveler safety (Cp1) is the most important criterion among the eight
identified criteria that users could consider when selecting shared mobility to make a trip, with
a weight = 0.1781. It is not surprising since criterion Cp1 is the level of safety of
individuals during the trip, such as the rate of accidents, harassment, assault, and theft. The
second most important criterion for users is cost (Cp7), which is the expenses for shared
mobility usage. The third most important criterion is accessibility (Cp3), which is the ease of
access, the availability of a shared vehicle, and proximity to the location of the parked shared
vehicle. Notably, these three criteria are related to the services' operations, which are at least
partially under the operator's control. The criterion comfort (Cp6), including the vehicle
characteristics that make people feel comfortable during the trip, and the criterion operational
speed (Cp2), which is the average velocity a shared mobility system overpasses, are the fourth
and fifth most important criteria, respectively. The operational speed is surprisingly less
important than all previous criteria. This interesting result questions the standard approach in
modeling modal choices whenever such services are considered since one of the key exogenous
variables is usually the travel time.
The sixth most important criterion for users is user-friendliness (Cp4), which means being easy
for beginners to learn and use and providing travel information in the app. Also, from the users'
point of view, the seventh most important criterion is the possibility of carrying items (Cp8),
0.2916 0.2756 0.2606 0.0832
0
0.5
Criteria Weight
Criteria
Operators' View (Shared Mobility)
Co1 Co2 Co3 Co5

231
which means carrying luggage or bags or shopping items in a shared vehicle. For instance,
people can carry their luggage by shared car, but not by scooter-sharing. The image (Cp5) is
the least important criterion for users, which is the image of a shared mobility system in the
eyes of the person. Also, Table 72 shows that criterion Cp1 is about 2.57 times more important
than criterion Cp5.
Furthermore, Figure 108 demonstrates the credal ranking of the criteria from the users'
point of view (for shared mobility services) and the assigned CL. It establishes that Cp1 is
certainly the most important criterion. As mentioned in section 3.2.7.6.4 of Chapter 3, when
the threshold value is 50 and the CL is above 80, it can be stated that one criterion is definitely
more important than another. It can be seen that the difference in importance among different
criteria is almost always confirmed, with the partial exception of cost versus accessibility,
comfort versus speed, and speed versus user friendliness for shared mobility users.
Figure 108: Credal ranking of criteria from users’ view for shared mobility services.
In summary, to better understand users' viewpoint on the impact of factors on their shared
mobility use, the weight of the three most important criteria, which are traveler safety (Cp1),
cost (Cp7), and accessibility (Cp3), respectively, and the weight of the least important criterion,
which is the image (Cp5), is demonstrated in Figure 109.

232
Figure 109: The weight of the least important criterion and the three most important criteria
(from users' perspective of shared mobility services).
6.2.4 Group weight of non-users of shared mobility services (as a whole, not
for a specific shared mobility service)
The optimal non-users’ group weights of the criteria for shared mobility services are
determined in Table 73.
Table 73: Non-users’ group weights of the criteria for shared mobility services.
Criteria
Weights
Ranking of criteria
Cp1. Traveler safety
0.1802
1
Cp2. Operational speed
0.1205
5
Cp3. Accessibility
0.1303
3
Cp4. User-friendliness
0.1267
4
Cp5. Image
0.0728
8
Cp6. Comfort
0.1179
6
Cp7. Cost
0.1433
2
Cp8. Possibility of carrying items
0.1083
7
Turning the attention to non-users, Table 73 shows that the three most important criteria are
still traveler safety, cost, and accessibility. However, user-friendliness is now coming up to the
fourth position, which underlines the importance of such a factor to increase the penetration of
shared mobility services and, at the same time, identifies the most important barrier to
achieving this goal. Conversely, the importance of comfort for non-user is slightly diminished
compared to other criteria.
Figure 110 establishes the credal ranking of the criteria from the no’-users' perspective
(for shared mobility services) and the assigned CL. It determines that Cp1 is definitely the most
important criterion. As mentioned in section 3.2.7.6.4 of Chapter 3, when the threshold value
is 50 and the CL is above 80, it can be mentioned that one criterion is absolutely more important
than another. It can be stated that the difference in importance between the different criteria is
almost always confirmed, with the minor exceptions of accessibility versus user-friendliness,
user-friendliness versus speed, and speed versus comfort for non-users.
0.1781 0.1437 0.1385 0.0694
0
0.2
Criteria Weight
Criteria
Users' View (Shared Mobility)
Cp1 Cp7 Cp3 Cp5

233
Figure 110: Credal ranking of criteria from non-users’ view for shared mobility services.
In summary, to better understand the viewpoint of non-users on the impact of factors on their
shared mobility usage, the weight of the three most important criteria, which are traveler safety
(Cp1), cost (Cp7), and accessibility (Cp3), respectively, and the weight of the least important
criterion, which is the image (Cp5), are given in Figure 111.
Figure 111: The weight of the least important criterion and the three most important criteria
(from the perspective of non-users of shared mobility services).
6.2.5 Similarities and differences between the four types of shared mobility
stakeholders (as a whole, not for a specific shared mobility service)
In this study, criteria (specified for each stakeholder) are compared by stakeholders in order to
recognize their viewpoints on the importance of each criterion. In the literature, some research
has only focused on the importance of some of these criteria. However, in this study, these
criteria are ranked and compared with each other to specify the importance of each criterion
compared with other criteria from each stakeholder's standpoint. Additionally, most studies
have worked on user perceptions only. However, in this research, the criteria related to each of
the four stakeholders have been identified. Therefore, this section helps to know how different
stakeholders score the importance of the comparison factors related to themselves (the third
research question mentioned in Chapter 1).
0.1802 0.1433 0.1303 0.0728
0
0.2
Criteria Weight
Criteria
Non-users' View (Shared Mobility)
Cp1 Cp7 Cp3 Cp5

234
Also, some criteria are compared by more than one stakeholder. Hence, the importance
of those criteria can be compared from the viewpoint of different stakeholders to distinguish
their views on each criterion. This contributes to knowing the perceptions of different groups
of shared mobility stakeholders about the importance of one criterion (related to the first
research question mentioned in Chapter 1).
The main shared mobility services shareholders, their criteria (related to each
stakeholder), and their corresponding weight are given in Table 74. This indicates the
importance of each criterion (specified for each stakeholder) compared to other criteria
determined by each stakeholder. It is important to mention that the corresponding weights of
government members, operators, users, and non-users are indicated by the “”, “”, “”
and “” respectively. Relevant criteria are considered the same for users and non-user
stakeholders, so their perceptions can be better compared. Similarly, some criteria are also
repeated in other groups. In this regard, the average number of trips per vehicle per day is an
important criterion for operators and government members stakeholders. Besides, operational
speed is an important criterion for operators, users, and non-users.
Table 74: Stakeholders, criteria, and related weights.
Criteria for government
members

Criteria for
operators

Criteria for
users

Criteria for
non-users

Cg1. Average number of
trips per vehicle per day
0.1488
Co1. Vehicle
utilization rate
0.2916
Cp1. Traveler
safety
0.1781
Cp1. People
safety
0.1802
Cg2. Greenhouse gases
(GHGs)
0.1528
Co2. Usage fees
0.2756
Cp2.
Operational
speed
0.1229
Cp2.
Operational
speed
0.1205
Cg3. Parking issues
0.1372
Co3. Average
number of trips
per vehicle per
day
0.2606
Cp3.
Accessibility
0.1385
Cp3.
Accessibility
0.1303
Cg4. Emission of
pollutants
0.1996
Co4.
Operational
speed
0.0890
Cp4. User-
friendliness
0.1171
Cp4. User-
friendliness
0.1267
Cg5. Integration of the
shared mobility service
with public transport
0.2505
Co5. The life
span of the
vehicle
0.0832
Cp5. Image
0.0694
Cp5. Image
0.0728
Cg6. Vehicle fee
0.1111
-
-
Cp6. Comfort
0.1260
Cp6. Comfort
0.1179
-
-
-
-
Cp7. Cost
0.1437
Cp7. Cost
0.1433
-
-
-
-
Cp8. Possibility
of carrying items
0.1042
Cp8.
Possibility of
carrying items
0.1083
As seen in Table 74, the average number of trips per vehicle per day is the government's fourth
most important criterion and the third most important one for the operators. In this regard, it is
worth stating that the importance of the criterion the average number of trips per vehicle per
day is 1.75 times higher for shared mobility operators than for government members. The
importance of this factor is more for operators than for government members. Besides, the
importance of the criterion operational speed is about 1.37 times higher for shared mobility
users and non-users than for operators.
Furthermore, Figure 112 reveals the weight percentage of the criteria corresponding
with the users and non-users of shared mobility services, which helps to understand their views
better. Interestingly, users and non-users of shared mobility services have a similar view on the
importance of all criteria. As listed in Tables 72 and 73, their three most important criteria are

235
traveler safety (Cp1), cost (Cp7), and accessibility (Cp3), respectively. However, as seen in
Table 72, the fourth and sixth most important criteria for users are comfort (Cp6) and user-
friendliness (Cp4), respectively. Conversely, as shown in Table 73, for non-users, the fourth
and sixth most important criteria are Cp4 and Cp6, respectively. Hence, compared to non-users,
shared mobility users give more importance to criterion Cp6 and less to criterion Cp4. Finally,
it is important to state that the user and non-users pay the least attention to the criterion
possibility of carrying items (Cp8) and criterion image (Cp5).
Figure 112: Importance of criteria based on users and non-users stakeholders.
6.2.6 Perception analysis
In this section, each stakeholder's (users and non-users) perception of the overall value of each
shared mobility service can be calculated using Eq. (2) presented in section 3 of Chapter 3 since
the weight assigned to the criterion and indicator value (score) of each criterion is determined.
As shown in Eq.2 in section 3 of Chapter 3, to calculate the stakeholder's perception of the
overall value of each shared mobility service, the first step is to multiply each criterion's
indicator value (score) by the weight (assigned by the stakeholder) of the criterion. Then it adds
all the results together. The higher the stakeholder's perception of the overall value of a type of
shared mobility service (compared to other types of shared mobility services), the greater the
stakeholder's preference for that type of shared mobility service (compared to other types of
shared mobility services). The analysis of the users’ and non-users’ perceptions of the overall
value of each shared mobility service is first reported together in subsection 5.5.6.1 since, as it
will be later explained, normalization of the indicator values is not required in these cases, as
described in section 3 of Chapter 3.
0.00 5.00 10.00 15.00 20.00
Cp1
Cp2
Cp3
Cp4
Cp5
Cp6
Cp7
Cp8
Weight Percentage
Criteria
Shared Mobility
Non-users Users

236
6.2.6.1 Perception analysis of users and non-users of shared mobility services (as a whole,
not for a specific shared mobility service)
Table 75 report the scores , i.e., the indicator values expressed by both users and non-users
of each shared mobility service were obtained from the above-described survey. Differences
between the two groups are determined as well. All scores are based on a 7-point scale;
therefore, the closer any indicator is to 7, the better the related shared mobility service performs
on that specific criterion. For instance, for criterion cost, 1 means very expensive, and 7 means
very cheap. Also, concerning, i.e., the possibility of carrying items, car-sharing is obviously
better assessed than bike-sharing and scooter-sharing. As expected, scores from users are
generally higher than the corresponding scores of non-users, with the only exception of the cost
of scooter-sharing, which is probably pointing to an underestimation of the monetary costs of
using such service by those that have no experience. Interestingly, accessibility and comfort
show the widest gap between users and non-users.
Table 75: Scores  obtained from users and non-users of each shared mobility service.
Criterion
Car-sharing services
Bike-sharing services
Scooter-sharing services
Users
Non-users
Diff.
Users
Non-users
Diff.
Users
Non-users
Diff.
Cp1. Traveler safety
5.40
4.94
0.46
4.31
3.96
0.35
3.18
3.09
0.09
Cp2. Operational speed
5.24
5.04
0.20
4.56
4.29
0.27
4.64
4.05
0.59
Cp3. Accessibility
5.07
4.53
0.54
5.09
4.22
0.87
5.16
4.45
0.71
Cp4. User-friendliness
5.11
4.60
0.51
4.91
4.42
0.49
4.91
4.49
0.42
Cp5. Image
5.38
4.95
0.43
4.82
4.36
0.46
4.69
4.11
0.58
Cp6. Comfort
5.36
4.65
0.71
4.53
3.96
0.57
3.84
3.15
0.69
Cp7. Cost
3.76
3.75
0.01
4.29
4.15
0.14
3.80
3.91
-
0.11
Cp8. Possibility of
carrying items
5.47
5.20
0.27
3.07
2.71
0.36
2.58
2.16
0.42
The next step is to calculate the perceived value of each alternative according to Eq.2 in section
3 of Chapter 3, multiplying each weight reported in the second columns of Tables 72 and 73
(for users and non-users, respectively) by the corresponding scores reported in Table 75. It
should be stated that since all scores from Table 75 have the same unit or scale ([1-7]), there is
no need to normalize them, thus 
 = , i, j. Results are reported in Tables 76 and 77,
respectively, for users and non-users, while the last row of each table represents the overall
value of each service V1, V2, and V3. Relative changes in % of the perceived value of bike-
sharing and scooter-sharing compared to car-sharing are reported in brackets. The higher the
users’ or non-users’ perceptions of the overall value of a type of shared mobility service
(compared to other types of shared mobility services), the greater the users’ or non-users’
preference for that type of shared mobility service (compared to other types of shared mobility
services).

237
Table 76: Perception of the value of each shared mobility service for users.
Users
Shared mobility services (% change compared to car-sharing)
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Perceived value
Perceived value
Perceived value
Cp1. Traveler safety
0.9617
0.7676 (-20%)
0.5664 (-41%)
Cp2. Operational speed
0.6440
0.5604 (-13%)
0.5703 (-11%)
Cp3. Accessibility
0.7022
0.7050 (0%)
0.7147 (2%)
Cp4. User-friendliness
0.5984
0.5750 (-4%)
0.5750 (-4%)
Cp5. Image
0.3734
0.3345 (-10%)
0.3255 (-13%)
Cp6. Comfort
0.6754
0.5708 (-15%)
0.4838 (-28%)
Cp7. Cost
0.5403
0.6165 (14%)
0.5461 (1%)
Cp8. Possibility of carrying items
0.5700
0.3199 (-44%)
0.2688 (-53%)
Vi
5.0654
4.4497 (-12%)
4.0506 (-20%)
Table 77: Perception of the value of each shared mobility service for non-users.
Non-users
Shared mobility services (% change compared to car-sharing)
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Perceived value
Perceived value
Perceived value
Cp1. Traveler safety
0.8902
0.7136 (-20%)
0.5568 (-37%)
Cp2. Operational speed
0.6073
0.5169 (-15%)
0.4880 (-20%)
Cp3. Accessibility
0.5903
0.5499 (-7%)
0.5798 (-2%)
Cp4. User-friendliness
0.5828
0.5600 (-4%)
0.5689 (-2%)
Cp5. Image
0.3604
0.3174 (-12%)
0.2992 (-17%)
Cp6. Comfort
0.5482
0.4669 (-15%)
0.3714 (-32%)
Cp7. Cost
0.5374
0.5947 (11%)
0.5603 (4%)
Cp8. Possibility of carrying items
0.5632
0.2935 (-48%)
0.2339 (-58%)
Vi
4.6798
4.0129 (-14%)
3.6583 (-22%)
As seen in Table 76, the users’ perception of the overall value of car-sharing (5.0654) is higher
than their perception of the overall value of bike-sharing (4.4497) and scooter-sharing (4.0506).
Similarly, Table 77 shows that the non-users' perception of the overall value of car-sharing
(4.6798) is higher than their perception of the overall value of bike-sharing (4.0129) and
scooter-sharing (3.6583). Therefore, based on the analysis of the eight criteria examined in this
study, car-sharing services are preferred by both users and non-users. Having a closer look at
the different patterns related to the contribution of each criterion to the overall value of one
alternative, it is not surprising to note that cost is the only one that gives the lowest contribution
to choosing car-sharing compared to its influence on choosing usually cheaper scooter-sharing
and bike-sharing services (in line with the scores in Table 75), as indicated by the positive
percent changes shown in the last two columns of the third last row of Tables 76 and 77.
Because on the 7-point survey for criterion cost, 1 means very expensive and 7 means very
cheap, car-sharing receives a lower score for this measure than bike-sharing and scooter-
sharing, which leads to a lower perceived value for the criterion cost of car-sharing. Also, bike-
sharing and scooter-sharing accessibility give a larger contribution to the value of these two
services for their users, while the opposite is true for non-users. Finally, scooter-sharing speed
is much less appreciated by non-users than by users. Note that this latter gap, embedding the
weights of each criterion according to Eq.2 in section 3 of Chapter 3, is relatively wider than
the average scores of the two groups related to scooter-sharing speed reported in Table 75.
6.2.7 Sensitivity analysis and scenarios
In this section, some scenarios are carried out to increase the use of bike-sharing and scooter-
sharing. In this regard, it is important to increase the motivation of users and non-users to make

238
a trip by bike-sharing and scooter-sharing services (compared to car-sharing services), as
mentioned in sub-section 6.2.7.1. Sensitivity analysis can be performed to evaluate the
scenarios that achieve this purpose.
6.2.7.1 Sensitivity analysis and scenario for users and non-users of shared mobility
services (as a whole, not for a specific shared mobility service)
For users and non-users groups, it should be noted that the indicator value of criterion cost
(Cp7) of bike-sharing and scooter-sharing is higher than that of car-sharing services, which
shows that from the point of view of users and non-users, the price of using bike-sharing and
scooter-sharing services is lower than car-sharing services. Besides, the indicator value of
criterion accessibility (Cp3) of bike-sharing and scooter-sharing is higher than that of car-
sharing services, which indicates that in the eyes of users, the accessibility of bike-sharing and
scooter-sharing is more than car-sharing. In this study, the indicator values of the criteria are
changed for both users and non-users so that it will be revealed if there is a difference.
Therefore, there is no need to change the indicator value of the criteria Cp7 and Cp3 for bike-
sharing and scooter-sharing because bike-sharing and scooter-sharing have a better situation
than car-sharing in terms of criterion Cp7 from the users' and non-users' standpoints, and
criterion Cp3 from the users' point of view.
It should also be stated that the change in the value of some criteria cannot be easily controlled
and analyzed in practice. In this regard, the change in the average velocity that a shared mobility
system overpasses (criterion operational speed (Cp2)) cannot be easily controlled. The
indicator value of the rest criteria, comprising the criteria user-friendliness (Cp4) and image
(Cp5), can be changed and used for the scenario.
6.2.7.1.1 Scenario for users and non-users groups: providing higher safety, higher
comfort, more user-friendly systems, a better image in the eyes of the public, and a better
possibility to carry items in bike-sharing and scooter-sharing services
In this scenario, bike-sharing and scooter-sharing services can provide higher safety (Cp1),
higher comfort (Cp6), more user-friendly systems (Cp4), a better image in the eyes of the public
(Cp5), and a better possibility to carry items (Cp8) in a way that users and non-users feel that
these features in these services are similar to these features in car-sharing services. To do this,
the indicator value of these criteria of bike-sharing and scooter-sharing services is set equal to
that of car-sharing services because car-sharing has a better situation than bike-sharing and
scooter-sharing in terms of these criteria from the users' and non-users' standpoints.
Tables 78 to 83 show the new calculation for users' and non-users' perceptions of the overall
value of each shared mobility service. The criteria (as well as the corresponding numbers) are
written in italics, and the bold font in Tables 78 to 83 has been changed compared to Tables 75
to 77.
Table 78: New indicator values for users' perception of the overall value of each shared
mobility service.
Users
Weight
Shared mobility services
Units
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Indicator value
Indicator value
Indicator value
Cp1. Traveler safety
0.1781
5.40
5.40
5.40
[1-7]

239
Cp2. Operational speed
0.1229
5.24
4.56
4.64
[1-7]
Cp3. Accessibility
0.1385
5.07
5.09
5.16
[1-7]
Cp4. User-friendliness
0.1171
5.11
5.11
5.11
[1-7]
Cp5. Image
0.0694
5.38
5.38
5.38
[1-7]
Cp6. Comfort
0.1260
5.36
5.36
5.36
[1-7]
Cp7. Cost
0.1437
3.76
4.29
3.80
[1-7]
Cp8. Possibility of carrying items
0.1042
5.47
5.47
5.47
[1-7]
Table 79: New indicator values for non-users' perception of the overall value of each shared
mobility service.
Non-users
Weight
Shared mobility services
Units
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Indicator value
Indicator value
Indicator value
Cp1. Traveler safety
0.1802
4.94
4.94
4.94
[1-7]
Cp2. Operational speed
0.1205
5.04
4.29
4.05
[1-7]
Cp3. Accessibility
0.1303
4.53
4.22
4.45
[1-7]
Cp4. User-friendliness
0.1267
4.60
4.60
4.60
[1-7]
Cp5. Image
0.0728
4.95
4.95
4.95
[1-7]
Cp6. Comfort
0.1179
4.65
4.65
4.65
[1-7]
Cp7. Cost
0.1433
3.75
4.15
3.91
[1-7]
Cp8. Possibility of carrying
items
0.1083
5.20
5.20
5.20
[1-7]
Table 80: New perception of the overall value of each shared mobility service analysis
results for users.
Users
Shared mobility services (% change compared to car sharing)
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Indicator value
Indicator value
Indicator value
Cp1. Traveler safety
0.9617
0.9617 (0%)
0.9617 (0%)
Cp2. Operational speed
0.6440
0.5604 (-13%)
0.5703 (-11%)
Cp3. Accessibility
0.7022
0.7050 (0%)
0.7147 (2%)
Cp4. User-friendliness
0.5984
0.5984 (0%)
0.5984 (0%)
Cp5. Image
0.3734
0.3734 (0%)
0.3734 (0%)
Cp6. Comfort
0.6754
0.6754 (0%)
0.6754 (0%)
Cp7. Cost
0.5403
0.6165 (14%)
0.5461 (1%)
Cp8. Possibility of carrying items
0.5700
0.5700 (0%)
0.5700 (0%)
Sum
5.0653
5.0607 (0%)
5.0098 (-1%)
Table 81: New perception of the overall value of each shared mobility service analysis
results for non-users.
Non-users
Shared mobility services (% change compared to car sharing)
Car-sharing services
Bike-sharing services
Scooter-sharing services
Criterion
Indicator value
Indicator value
Indicator value
Cp1. Traveler safety
0.8902
0.8902 (0%)
0.8902 (0%)
Cp2. Operational speed
0.6073
0.5169 (-15%)
0.4880 (-20%)
Cp3. Accessibility
0.5903
0.5499 (-7%)
0.5798 (-2%)
Cp4. User-friendliness
0.5828
0.5828 (0%)
0.5828 (0%)
Cp5. Image
0.3604
0.3604 (0%)
0.3604 (0%)
Cp6. Comfort
0.5482
0.5482 (0%)
0.5482 (0%)
Cp7. Cost
0.5374
0.5947 (11%)
0.5603 (4%)
Cp8. Possibility of carrying items
0.5632
0.5632 (0%)
0.5632 (0%)
Sum
4.6797
4.6063 (-2%)
4.5729 (-2%)
As seen in Table 80, increasing the indicator values of traveler safety (Cp1), user-friendliness
(Cp4), image (Cp5), comfort (Cp6), and the possibility of carrying items (Cp8) of bike-sharing
and scooter-sharing (to be equal to those of car-sharing) leads to a change in user's perception

240
of the overall value of bike-sharing (from -12% (shown in Table 76) to 0%) and scooter-sharing
services (from -20% (indicated in Table 77) to -1%) (compared to car-sharing services). Also,
Table 81 establishes that raising the indicator values of these criteria of bike-sharing and
scooter-sharing (to be equal to those of car-sharing) causes a change in non-users' perception
of the overall value of bike-sharing (from -14% (indicated in Table 77) to -2%) and scooter-
sharing services (from -22% (shown in Table 77) to -2%) (compared to car-sharing services).
Furthermore, for better understanding, Tables 82 and 83 systematically explore how results are
affected when only a subset of the five criteria mentioned above are changed. Therefore, Table
82 lists 83 (number of possible scenarios = -1=-1=32-1=31, where n is the number of
criteria selected to be increased for scenarios, which is 5) possible scenarios for users and non-
users groups, respectively, where scenario Cp1_4_5_6_8 (people’ safety, user-friendliness,
image, comfort, and possibility of carrying items) (increasing indicator value of criterion 1,
criterion 4, criterion 5, criterion 6, and criterion 8 of bike-sharing or scooter-sharing services,
so that be equal to those of car-sharing services) is the previously considered one. This scenario
and its corresponding numbers are in bold and italic font in Tables 82 and 83. Hence, this
scenario obviously leads to the best results to increase both uses of bike-sharing and scooter-
sharing by both users and non-users compared to the current situation because this scenario
includes all the increased criteria. Further, the rank of scenarios in situations where the purpose
is to increase the use of bike-sharing (compared to the use of car-sharing) and also the rank of
scenarios in cases where the aim is to raise the usage of scooter-sharing (compared to the use
of car-sharing) are presented for increasing the use of users and non-users in Table 82 and
Table 83, respectively.
Table 82: Current situation and possible scenarios for the users’ perception of the overall
value of each shared mobility service and the corresponding scenarios ranks (as a whole, not
for a specific shared mobility service).
Possible scenarios (changed
criteria)
Users’ perception of the overall value of each shared
mobility service (% change compared to car sharing)
Rank of scenarios (increasing the
shared mobility service use of
users)
Car-sharing
Bike-sharing
Scooter-sharing
Rank of
scenarios
(increasing
bike-
sharing
use of
users)
Rank of scenarios
(increasing
scooter-sharing
use of users)
(Current situation)
5.0653
4.4496 (-12%)
4.0505 (-20%)
-
-
Cp1 (people’s safety)
5.0653
4.6437 (-8%)
4.4458 (-12%)
24
20
Cp1_4 (people’s safety and user-
friendliness)
5.0653
4.6672 (-8%)
4.4693 (-12%)
23
19
Cp1_5 (people’s safety and image)
5.0653
4.6826 (-8%)
4.4937 (-11%)
22
18
Cp1_6 (people’s safety and
comfort)
5.0653
4.7483 (-6%)
4.6374 (-8%)
17
12
Cp1_8 (traveler safety and
possibility of carrying items)
5.0653
4.8938 (-3%)
4.7470 (-6%)
8
8
Cp1_4_5 (people’s safety, user-
friendliness, and image)
5.0653
4.7060 (-7%)
4.5171 (-11%)
20
17
Cp1_4_6 (people’s safety, user-
friendliness, and comfort)
5.0653
4.7717 (-6%)
4.6608 (-8%)
15
11
Cp1_4_8 (people’s safety, user-
friendliness, and possibility of
carrying items)
5.0653
4.9172 (-3%)
4.7704 (-6%)
7
7
Cp1_4_5_6 (people’s safety, user-
friendliness, image, and comfort)
5.0653
4.8106 (-5%)
4.7087 (-7%)
12
9

241
Possible scenarios (changed
criteria)
Users’ perception of the overall value of each shared
mobility service (% change compared to car sharing)
Rank of scenarios (increasing the
shared mobility service use of
users)
Car-sharing
Bike-sharing
Scooter-sharing
Rank of
scenarios
(increasing
bike-
sharing
use of
users)
Rank of scenarios
(increasing
scooter-sharing
use of users)
Cp1_4_5_8 (people’s safety, user-
friendliness, image, and possibility
of carrying items)
5.0653
4.9561 (-2%)
4.8183 (-5%)
5
5
Cp1_4_6_8 (people’s safety, user-
friendliness, comfort, and
possibility of carrying items)
5.0653
5.0218 (-1%)
4.9619 (-2%)
3
3
Cp1_5_6 (people’s safety, image,
and comfort)
5.0653
4.7872 (-5%)
4.6852 (-8%)
14
10
Cp1_5_8 (people’s safety, image,
and possibility of carrying items)
5.0653
4.9327 (-3%)
4.7949 (-5%)
6
6
Cp1_5_6_8 (people’s safety, image,
comfort, and possibility of carrying
items)
5.0653
5.0373 (-1%)
4.9864 (-2%)
2
2
Cp1_6_8 (people’s safety, comfort,
and possibility of carrying items)
5.0653
4.9984 (-1%)
4.9385 (-3%)
4
4
Cp1_4_5_6_8 (people’s safety, user-
friendliness, image, comfort, and
possibility of carrying items)
5.0653
5.0607 (0%)
5.0098 (-1%)
1
1
Cp4 (user-friendliness)
5.0653
4.4730 (-12%)
4.0739 (-20%)
31
31
Cp4_5 (user-friendliness and
image)
5.0653
4.5119 (-11%)
4.1218 (-19%)
29
29
Cp4_6 (user-friendliness and
comfort)
5.0653
4.5776 (-10%)
4.2654 (-16%)
27
27
Cp4_8 (user-friendliness and
possibility of carrying items)
5.0653
4.7231 (-7%)
4.3750 (-14%)
19
23
Cp4_5_6 (user-friendliness, image,
and comfort)
5.0653
4.6165 (-9%)
4.3133 (-15%)
25
25
Cp4_5_8 (user-friendliness, image,
and possibility of carrying items)
5.0653
4.7620 (-6%)
4.4229 (-13%)
16
21
Cp4_6_8 (user-friendliness,
comfort, and possibility of carrying
items)
5.0653
4.8277 (-5%)
4.5665 (-10%)
11
15
Cp4_5_6_8 (user-friendliness,
image, comfort, and possibility of
carrying items)
5.0653
4.8666 (-4%)
4.6144 (-9%)
9
13
Cp5 (image)
5.0653
4.4885 (-11%)
4.0983 (-19%)
30
30
Cp5_6 (image and comfort)
5.0653
4.5931 (-9%)
4.2899 (-15%)
26
26
Cp4_8 (image and possibility of
carrying items)
5.0653
4.7386 (-6%)
4.3995 (-13%)
18
22
Cp4_5_8 (image, comfort, and
possibility of carrying items)
5.0653
4.8431 (-4%)
4.5910 (-9%)
10
14
Cp6 (comfort)
5.0653
4.5542 (-10%)
4.2420 (-16%)
28
28
Cp6_8 (comfort and possibility of
carrying items)
5.0653
4.8043 (-5%)
4.5431 (-10%)
13
16
Cp8 (possibility of carrying items)
5.0653
4.6997 (-7%)
4.3516 (-14%)
21
24
Table 83: Current situation and possible scenarios for the non-users’ perception of the overall
value of each shared mobility service and the corresponding scenarios ranks (as a whole, not
for a specific shared mobility service).
Possible scenarios (changed
criteria)
Non-users perception of the overall value of each
shared mobility service (% change compared to car
sharing)
Rank of scenarios (increasing the
shared mobility service use of non-
users)
Car-sharing
Bike-sharing
Scooter-sharing
Rank of
scenarios
(increasing
bike-sharing
use of non-
users)
Rank of scenarios
(increasing scooter-
sharing use of non-
users)
(Current situation)
4.6797
4.0129 (-14%)
3.6584 (-22%)
-
-
Cp1 (people’s safety)
4.6797
4.1895 (-10%)
3.9918 (-15%)
24
23

242
Possible scenarios (changed
criteria)
Non-users perception of the overall value of each
shared mobility service (% change compared to car
sharing)
Rank of scenarios (increasing the
shared mobility service use of non-
users)
Car-sharing
Bike-sharing
Scooter-sharing
Rank of
scenarios
(increasing
bike-sharing
use of non-
users)
Rank of scenarios
(increasing scooter-
sharing use of non-
users)
Cp1_4 (people’s safety and
user-friendliness)
4.6797
4.2123 (-10%)
4.0057 (-14%)
23
21
Cp1_5 (people’s safety and
image)
4.6797
4.2324 (-10%)
4.0529 (-13%)
22
19
Cp1_6 (people’s safety and
comfort)
4.6797
4.2708 (-9%)
4.1686 (-11%)
20
15
Cp1_8 (traveler safety and
possibility of carrying items)
4.6797
4.4592 (-5%)
4.3210 (-8%)
8
8
Cp1_4_5 (people’s safety,
user-friendliness, and image)
4.6797
4.2553 (-9%)
4.0668 (-13%)
21
17
Cp1_4_6 (people’s safety,
user-friendliness, and
comfort)
4.6797
4.2937 (-8%)
4.1825 (-11%)
18
13
Cp1_4_8 (people’s safety,
user-friendliness, and
possibility of carrying items)
4.6797
4.4820 (-4%)
4.3349 (-7%)
7
7
Cp1_4_5_6 (people’s safety,
user-friendliness, image, and
comfort)
4.6797
4.3366 (-7%)
4.2437 (-9%)
14
9
Cp1_4_5_8 (people’s safety,
user-friendliness, image, and
possibility of carrying items)
4.6797
4.5249 (-3%)
4.3961 (-6%)
5
5
Cp1_4_6_8 (people’s safety,
user-friendliness, comfort,
and possibility of carrying
items)
4.6797
4.5633 (-2%)
4.5118 (-4%)
3
3
Cp1_5_6 (people’s safety,
image, and comfort)
4.6797
4.3138 (-8%)
4.2298 (-10%)
16
11
Cp1_5_8 (people’s safety,
image, and possibility of
carrying items)
4.6797
4.5021 (-4%)
4.3821 (-6%)
6
6
Cp1_5_6_8 (people’s safety,
image, comfort, and
possibility of carrying items)
4.6797
4.5835 (-2%)
4.5590 (-3%)
2
2
Cp1_6_8 (people’s safety,
comfort, and possibility of
carrying items)
4.6797
4.5405 (-3%)
4.4978 (-4%)
4
4
Cp1_4_5_6_8 (people’s
safety, user-friendliness,
image, comfort, and
possibility of carrying items)
4.6797
4.6063 (-2%)
4.5729 (-2%)
1
1
Cp4 (user-friendliness)
4.6797
4.0357 (-14%)
3.6723 (-22%)
31
31
Cp4_5 (user-friendliness and
image)
4.6797
4.0787 (-13%)
3.7335 (-20%)
29
29
Cp4_6 (user-friendliness and
comfort)
4.6797
4.1171 (-12%)
3.8492 (-18%)
27
27
Cp4_8 (user-friendliness and
possibility of carrying items)
4.6797
4.3054 (-8%)
4.0016 (-14%)
17
22
Cp4_5_6 (user-friendliness,
image, and comfort)
4.6797
4.1600 (-11%)
3.9103 (-16%)
25
25
Cp4_5_8 (user-friendliness,
image, and possibility of
carrying items)
4.6797
4.3483 (-7%)
4.0627 (-13%)
13
18
Cp4_6_8 (user-friendliness,
comfort, and possibility of
carrying items)
4.6797
4.3867 (-6%)
4.1784 (-11%)
11
14
Cp4_5_6_8 (user-
friendliness, image, comfort,
and possibility of carrying
items)
4.6797
4.4297 (-5%)
4.2396 (-9%)
9
10
Cp5 (image)
4.6797
4.0558 (-13%)
3.7195 (-21%)
30
30
Cp5_6 (image and comfort)
4.6797
4.1372 (-12%)
3.8964 (-17%)
26
26
Cp4_8 (image and possibility
of carrying items)
4.6797
4.3255 (-8%)
4.0488 (-13%)
15
20

243
Possible scenarios (changed
criteria)
Non-users perception of the overall value of each
shared mobility service (% change compared to car
sharing)
Rank of scenarios (increasing the
shared mobility service use of non-
users)
Car-sharing
Bike-sharing
Scooter-sharing
Rank of
scenarios
(increasing
bike-sharing
use of non-
users)
Rank of scenarios
(increasing scooter-
sharing use of non-
users)
Cp4_5_8 (image, comfort,
and possibility of carrying
items)
4.6797
4.4069 (-6%)
4.2256 (-10%)
10
12
Cp6 (comfort)
4.6797
4.0942 (-13%)
3.8352 (-18%)
28
28
Cp6_8 (comfort and
possibility of carrying items)
4.6797
4.3639 (-7%)
4.1645 (-11%)
12
16
Cp8 (possibility of carrying
items)
4.6797
4.2826 (-8%)
3.9876 (-15%)
19
24
As can be seen in Tables 82 and 83, it is interesting that from the perspective of both users and
non-users, the best scenario (highest usage increase) for both bike-sharing and scooter-sharing
is scenario Cp1_4_5_6_8 (people’s safety, user-friendliness, image, comfort, and possibility of
carrying items), followed by scenarios Cp1_5_6_8 (people’ safety, image, comfort, and
possibility of carrying items), Cp1_4_6_8 (people’ safety, user-friendliness, comfort, and
possibility of carrying items), Cp1_6_8 (people’ safety, comfort, and possibility of carrying
items), Cp1_4_5_8 (people’ safety, user-friendliness, image, and possibility of carrying items),
Cp1_5_8 (people’ safety, image, and possibility of carrying items), Cp1_4_8, and Cp1_8
(traveler safety and possibility of carrying items). On the other hand, from the point of view of
both users and non-users, the worst scenario (least usage increase) for both bike-sharing and
scooter-sharing is scenario Cp4 (user-friendliness), followed by scenarios, Cp6 (comfort),
Cp4_5 (user-friendliness and image), Cp5 (image), Cp4_6 (user-friendliness and comfort),
Cp5_6 (image, and comfort), and Cp4_5_6 (user-friendliness, image, and comfort).

244
Chapter 7
Conclusions
This study aims to identify the gap between the needs, expectations, and views of different
stakeholders in car-sharing, bike-sharing, and scooter-sharing systems. To do this, this study
has two different parts. These parts are the analysis of each shared mobility service (separately)
and the analysis of shared mobility services (as a whole, not for a specific shared mobility
service). Analyses were carried out through the use of the Bayesian Best-Worst-Method
(Bayesian BWM), the state-of-the-art method in multi-criteria analyses.
In the analysis of each shared mobility service (separately), 12 sub-criteria are
compared by four different groups of stakeholders in order to understand their views on the
importance of each sub-criterion that people can consider in their decisions to use each shared
mobility service. Also, in the analysis of shared mobility services (as a whole, not for a specific
shared mobility service), each stakeholder rated the importance of specific criteria associated
with their specific role in shared mobility service. Hence, government members, operators, and
users/non-users rated three partially different sets of criteria. However, users and non-users
rated the same criteria to understand the gap between their perceptions.
This experimental design allowed some original contributions to the field of multi-
criteria analyses and Bayesian BWM applications. More in detail:
• Some studies in the literature have only worked on the importance of some of these 12
sub-criteria. However, in this study, all 12 sub-criteria are ranked and compared with
each other to determine their relative importance from each stakeholder's perspective.
• Three different shared mobility services are considered: car-sharing, bike-sharing, and
scooter-sharing. Therefore, this study helps to understand how one main-criterion/sub-
criterion can be of different importance across different shared mobility services.
• Most studies have worked on user perspectives only. However, in this study, these sub-
criteria are compared by four groups of stakeholders. Therefore, the importance of each
sub-criterion can be compared from the perspective of these four different stakeholders
to distinguish their views on each sub-criterion.

245
• It is also important to note that in the literature, sub-criteria service quality and safety,
environment-friendly system, and user-friendly have not been well studied. Hence, this
study also considers these sub-criteria to determine the stakeholders' views on them.
• By analyzing and comparing the similarities and differences (gaps) in the perspectives
of each shared mobility service stakeholder, suggestions for government members and
each shared mobility service operator are given to attract more users and non-users.
• Additionally, most studies have worked on users’ perceptions only. In contrast, the
criteria of this study encompass additional evaluation dimensions, including factors
associated with the role of operators and government members as stakeholders of
shared mobility services. Also, the perception of non-users is studied to determine the
difference between their views compared to users. Therefore, our results help to know
how different stakeholders score the importance of the comparison factors associated
with their role as shared mobility service stakeholders.
• More than one stakeholder assesses some criteria. Hence, the importance of those
criteria can be compared from the viewpoint of different stakeholders to distinguish
their views on each criterion. This contributes to knowing the perceptions of different
groups of shared mobility stakeholders about the importance of one criterion. Besides,
this study help to understand which shared mobility system is most appropriate to
implement according to users' and non-users' perceptions. Also, this study contributes
to presenting scenarios from the views of users and non-users groups to determine how
to increase the use of bike-sharing and scooter-sharing services compared to car-sharing
services from users' and non-users' perspectives.
Furthermore, the following two points can be mentioned for the methodological contribution
of this research.
• Joint consideration of Multi-Actor Multi-Criteria Analysis (MAMCA) and Bayesian
Best-Worst Method (BWM) for Perception-Based Analysis (PBA)
• Introducing the Confidence Level (CL) classification in the Credal Ranking (Bayesian
BWM) based on previous literature
From a methodological viewpoint, the above-mentioned Bayesian BWM is framed within a
Multi-Actor Multi-Criteria Analysis (MAMCA) since the latter is an appropriate method when
different stakeholders are involved. More specifically, the third step of the MAMCA is to
determine the main criteria and weights, which is done through a Perception-Based Analysis
(PBA) that implements a Bayesian BWM in the present research. This method is chosen since
it is the only one ensuring a very high quality of the computed weights while requiring a small
amount of data. This aspect is very important because some of the shareholders are members
of the government and operators, which are few in number. Other advantages of this method
include the combination of weight quality, fewer inconsistencies between criteria, fewer data
required to obtain highly reliable results, low equalizing bias, and average transparency of the
method.
Before calculating the optimal group weights by Bayesian BWM, the consistency of
the respondents was examined using the input-based approach, and acceptable ones (their

246
obtained global input-based consistency ratio is less than the input-based consistency ratio
thresholds) were considered. After eliminating pairwise comparisons with unacceptable
consistency ratios, different sample sizes can be obtained and utilized for different levels of the
model. Also, it is important to note that Bayesian BWM can provide much more information
than the original BWM. For example, Bayesian BWM can provide the credal ranking and
confidence level in the weight-directed graph. This helps to understand the importance
perceived by stakeholders of one criterion over other criteria.
In order to gather the required data, nine different surveys have been designed and
administered in our study area, namely the Turin metropolitan area in Italy. Data on operators
and government members were collected through phone calls to targeted contact points,
whereas for users and non-users, it was possible to resort to a panel maintained by a survey
company to have a representative sample of the population in the study area. In addition, online
surveys were administered to the panel members. Survey data are used to calculate the criteria
and sub-criteria weights to determine how the comparative criteria are rated in terms of
importance by different stakeholders of different shared mobility services. Hence, surveys help
to gain insights into how specific individuals or groups perceive specific aspects. In those
surveys administered to users and non-users of each shared mobility service, in addition to
BWM-related questions, questions about their routines, daily travel views, and socio-
demographic characteristics were asked as well.
The obtained data associated with the views of operators and members of the
government regarding some of the travel routines of users of each of the shared transportation
services shows that from the perspective of at least half of car-sharing users and government
members (who responded to the car-sharing survey), short-time trips (less than 30 min) can
induce people to use (or use more) car-sharing. However, trips beyond 30 min cannot do that.
On the other hand, none of the car-sharing operators agrees with the statement. This is an
example of the gap between the views of car-sharing operators, car-sharing users, and
government members (who responded to the car-sharing survey) about the effect of short-time
trips on car-sharing demand. More detailed results are in the remainder.
Key conclusions from the descriptive statistics of the collected data
Some of the important results obtained from the collected data associated with the routines and
daily travel views of users and non-users are as follows.
• The most common use of car-sharing is to perform a work-related activity in the city
center. However, most people are likely to use bike-sharing and scooter-sharing for
weekend activities.
• Concerning temporal patterns, car-sharing is mainly preferred during the off-peak
hours; however, bike-sharing and scooter-sharing are mostly preferred for peak hours.
• Non-users of car-sharing are not paying much attention to the potentialities of this
service for leisure travel. On the other hand, non-users of bike-sharing do not consider
the capacity of this service for non-recreational trips. Regarding scooter-sharing, it can
be mentioned that it has the potential to be used for both travel purposes.

247
• Increased comfort during travel is more important to male bike-sharing users than to
female users. On the other hand, the availability of the service near the user's
home/work and avoiding responsibilities related to maintenance and repairs are more
important for females than males as a motivation to use bike-sharing.
• Compared to female car-sharing users, male car-sharing users are more interested in
using the service only for non-leisure (going to work/school) trips. However, compared
to female bike-sharing users, male bike-sharing users are more inclined to use the
service only for leisure (e.g., visiting friends or shopping) trips. Concerning leisure-
only travel (e.g., visiting friends or shopping), female scooter-sharing users are keener
than male users.
The models' results can be divided into two parts: the conclusions from the analysis for each
shared mobility service (separately) and for shared mobility services (as a whole, not for a
specific shared mobility service). All results are reported concerning the main-criteria and sub-
criteria, introduced in Tables 33 to 36 of Chapter 4.
Key conclusions from the analysis results for each shared mobility service (separately):
car sharing
Some suggestions and policies derived from the similarities and differences between the four
types of car-sharing stakeholders are given as follows.
• Car-sharing operators should pay more attention to trip-related characteristics instead
of car-sharing characteristics in order to attract more users and non-users.
• Car-sharing operators and government members should pay attention to availability and
accessibility to satisfy car-sharing service users; government members should pay more
attention since they believe that availability and accessibility are the least important
criterion.
• Car-sharing operators can focus less on user-friendliness, service quality, and safety
(minimum safety required) to attract users and non-users.
• Operators should place more value on travel costs in their policies. Correspondingly,
operators can reduce the cost of car-sharing services to the public to attract non-users
and satisfy users. Also, due to more and easier access to free-floating car-sharing
services than one-way or two-way car-sharing services, the travel time and travel
distance of people traveling by free-floating car-sharing services can be less. Hence,
operators can offer free-floating car-sharing to attract non-users and encourage users to
use it more.
• Government members should note that they underestimate the importance of the
availability of car-sharing services and travel distance in car-sharing demand. However,
on the other hand, they overestimate the importance of the cost of car-sharing and trip
purposes in car-sharing demand.
Key conclusions from the analysis results for each shared mobility service (separately):
bike-sharing

248
Some suggestions and policies derived from the similarities and differences between the four
types of stakeholders of bike-sharing are presented as follows.
• Bike-sharing operators should place more value on trip-related characteristics instead
of bike-sharing characteristics to attract more individuals, especially non-users.
• Government members should be aware that they underestimate the importance of bike-
sharing characteristics, availability, and accessibility compared to people, especially
users. On the other hand, they overestimate the importance of trip-related
characteristics.
• Bike-sharing operators can pay less attention to environment-friendly issues and place
more value on trip-related characteristics, especially trip purpose and travel time, to
attract users and non-users.
• Bike-sharing operators should pay more attention to vehicle availability and
accessibility. In this regard, by switching from station-based bike-sharing to free-
floating bike-sharing, operators may attract more users and non-users because people
may have easier access to bike-sharing. Also, they do not need to ride a bike to reach a
particular station. Hence, their travel time and distance can be shorter, leading to more
bike-sharing users.
• Government members should be aware that they are underestimating the role of comfort
and environmental-friendly system in demand for bike-sharing. However, on the other
hand, they overestimate the role of travel time.
• Government members should realize that they underestimate the importance of safety
compared to non-users, and bike-sharing operators should pay more attention to service
quality to encourage users.
Key conclusions from the analysis results for each shared mobility service (separately):
scooter-sharing
Some suggestions and policies are offered from the similarities and differences between the
four types of scooter-sharing stakeholders.
• Government members should know that they underestimate trip-related characteristics
compared to non-users. However, on the other hand, they overestimate scooter-sharing
characteristics.
• More attention is required by scooter-sharing operators to scooter-sharing
characteristics to attract more users and non-users.
• Scooter-sharing operators can pay more attention to service availability than vehicle
availability and accessibility to encourage people to use scooter-sharing, especially
non-users.
• To attract more users, scooter-sharing operators need to focus more on travel comfort
and service quality.
• Scooter-sharing operators should pay more attention to travel costs, especially to raise
user engagement.
• In general, scooter-sharing operators should offer more comfort services and high-
quality scooter-sharing in high-demand locations at lower prices to increase demand.

249
• Government members should be aware that they underestimate travel time, travel
distance, departure time, and vehicle availability and accessibility compared to people,
especially users of scooter-sharing. However, on the other hand, they overestimate
travel costs, safety, environment-friendly system, and user-friendly.
Tables 84 and 85 summarize the above-listed suggestions for government members and
operators to pay more attention (+) (because they underestimate) or less attention (-) (because
they overestimate compared to users/non-users) to the main-criteria and sub-criteria,
respectively. For instance, Table 85 shows that because government members overestimate the
importance of travel time (compared to users/non-users), they can pay less attention (-) to it
(compared to now (if it has a role in their policy-making)). On the other hand, since bike-
sharing operators underestimate the importance of travel time (compared to users/non-users),
they should pay more attention (+) to it (compared to now).
Table 84: Suggestions for government members and operators to pay more attention (+)
(because they underestimate) or less attention (-) (because they overestimate) to the
importance of the main-criteria.
Shared Mobility Services
Main-criteria
Government members
Operators
Car-sharing
Trip-related Characteristics
(+)
Car-sharing characteristics
(-)
Bike-sharing
Trip-related Characteristics
(-)
(+)
Bike-sharing characteristics
(+)
(-)
Availability and accessibility
(+)
Scooter-sharing
Trip-related Characteristics
(+)
Scooter-sharing characteristics
(-)
(+)
Table 85: Suggestions for government members and operators to pay more attention (+)
(because they underestimate) or less attention (-) (because they overestimate) to the
importance of sub-criteria.
Shared mobility services
Sub-criteria
Government members
Operators
Car-sharing
Travel distance
(+)
Trip purpose
(-)
Travel cost
(-)
(+)
Safety
(-)
Service quality
(-)
User-friendly
(-)
Service availability
(+)
Vehicle availability and accessibility
(+)
(+)
Bike-sharing
Travel time
(-)
(+)
Trip purpose
(+)
Travel comfort
(+)
Safety
(+)
Service quality
(+)
Environment-friendly system
(+)
(-)
Scooter-sharing
Travel time
(+)
Travel distance
(+)
Departure time
(+)
Travel cost
(-)
(+)
Travel comfort
(+)
Safety
(-)
Service quality
(+)
Environment-friendly system
(-)
User-friendly
(-)
Service availability
(+)
Vehicle availability and accessibility
(+)
(-)

250
Key conclusions from the analysis results for each shared mobility service (separately):
Views of the four stakeholders related to different services
Some conclusions from the comparison between the views of the four stakeholders related to
different services are delivered as follows.
• Government members consider that trip purpose, travel cost, and vehicle availability
and accessibility are twice as important as some of the other criteria concerning car-
sharing services, whereas travel time, travel distance, service availability, and vehicle
availability and accessibility are prominent for them when dealing with scooter-sharing
services.
• Car-sharing operators consider that user-friendliness, service availability, and vehicle
availability and accessibility are twice as important as some other car-sharing criteria.
For the bike-sharing operator, environment-friendly systems, service availability,
vehicle availability, and accessibility are prominent.
• Car-sharing users believe that travel time, service availability, and vehicle availability
and accessibility are at least twice as important as some other car-sharing criteria. Also,
bike-sharing users believe that service availability, vehicle availability, and
accessibility are at least twice as important as some other bike-sharing criteria. In this
regard, scooter-sharing users believe that safety, vehicle availability, and accessibility
are at least twice as important as some other criteria concerning scooter-sharing.
Besides, users of all shared mobility services consider that vehicle availability and
accessibility factor is at least twice as important as the departure time.
• Non-users of car-sharing consider that service availability, vehicle availability, and
accessibility are at least twice more important than some other criteria. Also, both bike-
sharing and scooter-sharing non-users believe that travel time, service availability, and
vehicle availability and accessibility are at least twice more important than some of the
other criteria. Further, non-users of all shared mobility services consider service
availability, vehicle availability, and accessibility at least twice as important as travel
comfort, environment-friendly system, and user-friendliness.
Key conclusions from MAMCA analysis for shared mobility services (as a whole)
From the analysis of the weights, it was concluded that the average number of trips per vehicle
per day is more important for operators than for government members. Also, operational speed
is more important for users and non-users than for operators. Besides, in the eyes of users and
non-users, the shared mobility system should be (in order of importance): safe, low-cost, and
highly accessible to both attract non-users and encourage more users to use it. Moreover, the
scores (of the criteria) given by users are generally higher than those of non-users except for
the cost of scooter-sharing, which may indicate that non-users underestimate the travel cost of
scooter-sharing services. Finally, it is worth mentioning that the two least important criteria
affecting the choice of shared mobility service from both users' and non-users' points of view
are (in order of importance) the possibility of carrying items and the image.
Furthermore, from the perception analysis, it is clear that based on the analysis of the
eight criteria examined in this study, car-sharing services (compared to bike-sharing services

251
and scooter-sharing services) were preferred by users and non-users of shared transportation
services in Turin, Italy. Besides, the cost is the only criterion with the least contribution to the
choice of car-sharing services (compared to the other two shared mobility services) by both
users and non-users. This result is different from the results obtained from the analysis of
weights, from which it was concluded that the cost of travel is the second most important
criterion in choosing a shared transportation service. As people have stated in their scoring,
car-sharing services cost more than bike-sharing and scooter-sharing services, which makes up
the difference because car-sharing receives a lower score, leading to a lower perceived value
for this criterion.
It should be pointed out that the scooter-sharing service has the lowest priority among
the three shared transportation services for users and non-users. The most important reason is
that carrying fewer items with this service than car-sharing, and the service is also less safe and
comfortable. Besides, from the standpoint of users and non-users, bike-sharing services are less
preferred than car-sharing services due to less possibility of carrying items, safety, and comfort.
On the other hand, from the analysis of the weights, it was concluded that the possibility of
carrying items is one of the least important criteria. As users and non-users have noted in their
scoring, both scooter-sharing and bike-sharing have less possibility to carry things than car-
sharing, which causes the difference between the results of the weights analysis and perception
analysis. Besides, it should be stated that the lower operational speed of bike-sharing
(compared to car-sharing) contributes to its low preference, especially in the eyes of non-users.
In addition, it is interesting to mention that the criteria accessibility and comfort show the
greatest perception gap between users and non-users. Also, bike-sharing and scooter-sharing
accessibility can contribute more to the value of these two services for their users, while the
opposite is true for non-users. Finally, the speed of scooter-sharing is much less appreciated by
non-users than by users. Note that this gap, embedding the weights of each criterion, is
relatively wider compared to the average scores of the two groups related to scooter-sharing
speed. The sensitivity analysis and scenario for users and non-users of shared mobility services
(as a whole, not for a specific shared mobility service) demonstrate that from the perspective
of both users and non-users, the best scenario to have the greatest increase in use for both bike-
sharing and scooter-sharing is a scenario in which people's safety, user-friendliness, image,
comfort, and the possibility of carrying items are increased.
This study provides suggestions to operators and government members to show how
the importance of sub-criteria and main-criteria can increase users' engagement and attract non-
users to services. Also, it contributes to knowing how different stakeholders score the
importance of the comparison factors associated with their roles as stakeholders of shared
mobility services. Besides, these results shed light on the relative importance of a set of criteria
in choosing different mobility-sharing services for both its users and non-users. However,
results are not necessarily correlated to the actual market share of the service. Indeed, car-
sharing has the overall best value, but it serves fewer trips compared to bike-sharing in Turin.
This is because different considerations might arise when making the final choice at the trip
level. In other words, the above-presented methodology is not a tool to forecast travel behaviors
or market shares of different services but rather to gain a deeper understanding of the factors

252
that are stronger drivers of the choices, including those that cannot easily or readily be captured
by observed or even latent variables or psychological constructs.
Considering the limitations of this study and recommendations for future studies, the
data collection process could be done face-to-face with respondents in future research. In that
case, the input-based approach can be performed during the meeting so that respondents can
modify their answers instantly, leading to less excluded data. Also, a new combination of BWM
with other appropriate methods, such as the fuzzy best-worst multi-criteria group decision-
making method for the third step of MAMCA, can be used to compare related results with
those of this study. In addition, to determine the overall importance of each criterion from the
point of view of all stakeholders (simultaneously), stakeholders can be assigned weights (in the
third stage of MAMCA). This was not done in this study because it was not our aim. This can
indicate the importance of stakeholders in the decision-making process. Also, a hierarchical
criteria tree (in the third stage of MAMCA) can be prepared to show the stakeholders involved
with their goals and objectives. In this study, the criteria selection is based on the objectives of
the stakeholders involved and the considered alternatives (car-sharing, bike-sharing, and
scooter-sharing).

253
References
Acheampong, R. A., & Siiba, A. (2019). Modelling the determinants of car-sharing adoption intentions among
young adults: the role of attitude, perceived benefits, travel expectations and socio-demographic
factors. Transportation, 1-24.
Agenzia per la Mobilità Metropolitana e Regionale, 2015. IMQ 2013. Indagine sulla Mobilità delle Persone e
sulla Qualità dei Trasporti. Rapporto di sintesi sull’area metropolitana. Torino.
Ahillen, M., Mateo-Babiano, D., & Corcoran, J. (2016). Dynamics of bike sharing in Washington, DC and
Brisbane, Australia: Implications for policy and planning. International journal of sustainable
transportation, 10(5), 441-454.
Ahmad, W. N. K. W., Rezaei, J., Sadaghiani, S., & Tavasszy, L. A. (2017). Evaluation of the external forces
affecting the sustainability of oil and gas supply chain using Best Worst Method. Journal of cleaner
production, 153, 242-252.
Ahmadi, HB, Kusi-Sarpong, S., & Rezaei, J. (2017). Assessing the social sustainability of supply chains using
Best Worst Method. Resources, Conservation and Recycling , 126 , 99-106.
Akar, G., Fischer, N., & Namgung, M. (2013). Bicycling choice and gender case study: The Ohio State
University. International Journal of Sustainable Transportation, 7(5), 347-365.
Allem, J. P., & Majmundar, A. (2019). Are electric scooters promoted on social media with safety in mind? A
case study on Bird's Instagram. Preventive medicine reports, 13, 62-63.
Almannaa, M. H., Ashqar, H. I., Elhenawy, M., Masoud, M., Rakotonirainy, A., & Rakha, H. (2020). A
Comparative Analysis of E-Scooter and E-Bike Usage Patterns: Findings from the City of Austin,
TX. arXiv preprint arXiv:2006.04033.
Amirnazmiafshar, E., & Diana, M. (2022). A review of the socio-demographic characteristics affecting the
demand for different car-sharing operational schemes. Transportation Research Interdisciplinary
Perspectives, 14, 100616.
Anderson-Hall, K., Bordenkircher, B., O'Neil, R., & Scott, S. C. (2019). Governing micro-mobility: A nationwide
assessment of electric scooter regulations (No. 19-05267).
Axhausen, K. (2013). Mobility Y: The emerging travel patterns of generation Y. Institute for Mobility Research,
Zurich, Switzerland.
Axsen, J., & Sovacool, B. K. (2019). The roles of users in electric, shared and automated mobility
transitions. Transportation Research Part D: Transport and Environment, 71, 1-21.

254
Bachand-Marleau, J., Lee, B. H., & El-Geneidy, A. M. (2012). Better understanding of factors influencing
likelihood of using shared bicycle systems and frequency of use. Transportation Research
Record, 2314(1), 66-71.
Badeau, A., Carman, C., Newman, M., Steenblik, J., Carlson, M., & Madsen, T. (2019). Emergency department
visits for electric scooter-related injuries after introduction of an urban rental program. The American
journal of emergency medicine, 37(8), 1531-1533.
Baek, K., Lee, H., Chung, J. H., & Kim, J. (2021). Electric scooter sharing: How do people value it as a last-mile
transportation mode?. Transportation Research Part D: Transport and Environment, 90, 102642.
Bagloee, S. A., Sarvi, M., & Wallace, M. (2016). Bicycle lane priority: Promoting bicycle as a green mode even
in congested urban area. Transportation Research Part A: Policy and Practice, 87, 102-121.
Bai, S., & Jiao, J. (2020). Dockless E-scooter usage patterns and urban built environments: a comparison study of
Austin, TX, and Minneapolis, MN. Travel behaviour and society, 20, 264-272.
Bajracharya, L., Mulya, T., Purbasari, A., & Hwang, M. (2018, June). A Study on Cost-Effective and Eco-friendly
Bicycle Sharing System for Developing Countries. In International Conference on Information Science
and Applications (pp. 523-531). Springer, Singapore.
Baker, L. (2009). How to get more bicyclists on the road. Scientific American, 301, 28-29.
Balac, M., Ciari, F., & Axhausen, K. W. (2015). Car-sharing demand estimation: Zurich, switzerland, area case
study. Transportation Research Record, 2563(1), 10-18.
Baltes, M. R. (1996). Factors influencing nondiscretionary work trips by bicycle determined from 1990 US census
metropolitan statistical area data. Transportation research record, 1538(1), 96-101.
Barnes, G., & Krizek, K. (2005). Estimating bicycling demand. Transportation Research Record, 1939(1), 45-51.
Barth, M., & Shaheen, S. A. (2002). Shared-use vehicle systems: Framework for classifying carsharing, station
cars, and combined approaches. Transportation Research Record, 1791(1), 105-112.
Basch, C. H., Ethan, D., Rajan, S., Samayoa-Kozlowsky, S., & Basch, C. E. (2014). Helmet use among users of
the Citi Bike bicycle-sharing program: a pilot study in New York City. Journal of community
health, 39(3), 503-507.
Basch, C. H., Zagnit, E. A., Rajan, S., Ethan, D., & Basch, C. E. (2014). Helmet use among cyclists in New York
City. Journal of community health, 39(5), 956-958.
Bauman, A., Crane, M., Drayton, B. A., & Titze, S. (2017). The unrealized potential of bike share schemes to
influence population physical activity levels–a narrative review. Preventive medicine, 103, S7-S14.
Beck, S., Barker, L., Chan, A., & Stanbridge, S. (2020). Emergency department impact following the introduction
of an electric scooter sharing service. Emergency Medicine Australasia, 32(3), 409-415.
Becker, H., Ciari, F., & Axhausen, K. W. (2017a). Comparing car-sharing schemes in Switzerland: User groups
and usage patterns. Transportation Research Part A: Policy and Practice, 97, 17-29.
Becker, H., Ciari, F., & Axhausen, K. W. (2017b). Modeling free-floating car-sharing use in Switzerland: A spatial
regression and conditional logit approach. Transportation Research Part C: Emerging Technologies, 81,
286-299.

255
Benayoun, R., Roy, B., & Sussman, N. (1966). Manual de reference du programme electre. Note de synthese et
Formation, 25, 79.
Berge, S. H. Kickstarting Micromobility–a Pilot Study on e-Scooters (No. 1721/2019). 2019.
Bhat, C. R., Dubey, S. K., & Nagel, K. (2015). Introducing non-normality of latent psychological constructs in
choice modeling with an application to bicyclist route choice. Transportation Research Part B:
Methodological, 78, 341-363.
Bian, T., Hu, J., & Deng, Y. (2017). Identifying influential nodes in complex networks based on AHP. Physica
A: Statistical Mechanics and its Applications, 479, 422-436.
Bloom, M. B., Noorzad, A., Lin, C., Little, M., Lee, E. Y., Margulies, D. R., & Torbati, S. S. (2021). Standing
electric scooter injuries: Impact on a community. The American Journal of Surgery, 221(1), 227-232.
Bonnette, B. (2007). The Implementation of a Public-Use Bicycle Program in Philadelphia.
Bonyun, M., Camden, A., Macarthur, C., & Howard, A. (2012). Helmet use in BIXI cyclists in Toronto, Canada:
an observational study. BMJ open, 2(3).
Bordagaray, M., dell’Olio, L., Fonzone, A., & Ibeas, Á. (2016). Capturing the conditions that introduce systematic
variation in bike-sharing travel behavior using data mining techniques. Transportation research part C:
emerging technologies, 71, 231-248.
Borgnat, P., Abry, P., Flandrin, P., Robardet, C., Rouquier, J. B., & Fleury, E. (2011). Shared bicycles in a city:
A signal processing and data analysis perspective. Advances in Complex Systems, 14(03), 415-438.
Botti, L., & Peypoch, N. (2013). Multi-criteria ELECTRE method and destination competitiveness. Tourism
Management Perspectives, 6, 108-113.
Bowman, B. L., Vecellio, R. L., & Haynes, D. W. (1994). Strategies for increasing bicycle and pedestrian safety
and use. Journal of urban planning and development, 120(3), 105-114.
Brandstätter, G., Gambella, C., Leitner, M., Malaguti, E., Masini, F., Puchinger, J., ... & Vigo, D. (2016).
Overview of optimization problems in electric car-sharing system design and management. In Dynamic
perspectives on managerial decision making (pp. 441-471). Springer, Cham.
Brans, J. P., & Vincke, P. (1985). Note—A Preference Ranking Organisation Method: (The PROMETHEE
Method for Multiple Criteria Decision-Making). Management science, 31(6), 647-656.
Brans, J. P., Vincke, P., & Mareschal, B. (1986). How to select and how to rank projects: The PROMETHEE
method. European journal of operational research, 24(2), 228-238.
Brendel, A. B., Brennecke, J. T., & Nastjuk, I. (2018). Applying Econophysics in the Context of Carsharing-
Development of a Vehicle Relocation Algorithm and Decision Support System.
Bridgman, P.W. (1922). Dimensional Analysis. New Haven: Yale University Press.
Brispat, P. (2017). Perception Based Decision-making for Public Transport Investments. Master’s Thesis, Delft
University of Technology, Delft, The Netherlands.
Brook, D. (2004, January). Carsharing–start up issues and new operational models. In Transportation Research
Board Annual Meeting.

256
Brownson, A. B., Fagan, P. V., Dickson, S., & Civil, I. D. (2019). Electric scooter injuries at Auckland City
Hospital. NZ Med J, 132, 62-72.
Brunelli, M., & Rezaei, J. (2019). A multiplicative best–worst method for multi-criteria decision
making. Operations Research Letters, 47(1), 12-15.
Buck, D., & Buehler, R. (2012, January). Bike lanes and other determinants of capital bikeshare trips. In 91st
Transportation research board annual meeting.
Buehler, R., & Hamre, A. (2014). Economic benefits of capital bikeshare: A focus on users and businesses (No.
VT-2013-06). Mid-Atlantic Universities Transportation Center.
Buehler, R., & Pucher, J. (2011). Making public transport financially sustainable. Transport Policy, 18(1), 126-
138.
Buehler, R., & Pucher, J. (2012). Cycling to work in 90 large American cities: new evidence on the role of bike
paths and lanes. Transportation, 39(2), 409-432.
Burghard, U., & Dütschke, E. (2019). Who wants shared mobility? Lessons from early adopters and mainstream
drivers on electric carsharing in Germany. Transportation Research Part D: Transport and
Environment, 71, 96-109.
Burkhardt, J. E., & Millard-Ball, A. (2006). Who is attracted to carsharing?. Transportation Research
Record, 1986(1), 98-105.
Büttner, J., & Petersen, T. (2011). Optimizing bike sharing in European cities-a handbook.
Button, K., Frye, H., & Reaves, D. (2020). Economic regulation and E-scooter networks in the USA. Research in
transportation economics, 100973.
Calderón, F., & Miller, E. J. (2020). A literature review of mobility services: definitions, modelling state-of-the-
art, and key considerations for a conceptual modelling framework. Transport Reviews, 40(3), 312-332.
Campbell, A. A., Cherry, C. R., Ryerson, M. S., & Yang, X. (2016). Factors influencing the choice of shared
bicycles and shared electric bikes in Beijing. Transportation research part C: emerging technologies, 67,
399-414.
Cao, H. J., Liu, F., Li, C. B., & Liu, C. (2006). An integrated method for product material selection considering
environmental factors and a case study. In Materials Science Forum (Vol. 532, pp. 1032-1035). Trans
Tech Publications Ltd.
Carrese, S., d’Andreagiovanni, F., Giacchetti, T., Nardin, A., & Zamberlan, L. (2021). A beautiful fleet: optimal
repositioning in e-scooter sharing systems for urban decorum. Transportation Research Procedia, 52,
581-588.
Carroll, P., Caulfield, B., & Ahern, A. (2017). Examining the potential for car-shedding in the Greater Dublin
Area. Transportation Research Part A: Policy and Practice, 106, 440-452.
Cartenì, A., Cascetta, E., & de Luca, S. (2016). A random utility model for park & carsharing services and the
pure preference for electric vehicles. Transport Policy, 48, 49-59.
Caspi, O., Smart, M. J., & Noland, R. B. (2020). Spatial associations of dockless shared e-scooter
usage. Transportation Research Part D: Transport and Environment, 86, 102396.

257
Catalano, M., Lo Casto, B., & Migliore, M. (2008). Car sharing demand estimation and urban transport demand
modelling using stated preference techniques.
Caulfield, B., O'Mahony, M., Brazil, W., & Weldon, P. (2017). Examining usage patterns of a bike-sharing
scheme in a medium sized city. Transportation research part A: policy and practice, 100, 152-161.
Ceccato, R. (2020). Switching intentions towards car sharing (Doctoral dissertation, Politecnico di Torino).
Ceccato, R., & Diana, M. (2021). Substitution and complementarity patterns between traditional transport means
and car sharing: a person and trip level analysis. Transportation, 48(4), 1523-1540.
Celsor, C., & Millard-Ball, A. (2007). Where does carsharing work? Using geographic information systems to
assess market potential. Transportation Research Record, 1992(1), 61-69.
Cervero, R. (2003). City CarShare: First-year travel demand impacts. Transportation research record, 1839(1),
159-166.
Cervero, R. (2009). TOD and carsharing: A natural marriage. ACCESS Magazine, 1(35), 25-29.
Cervero, R., & Tsai, Y. (2004). City CarShare in San Francisco, California: second-year travel demand and car
ownership impacts. Transportation Research Record, 1887(1), 117-127.
Cervero, R., Golub, A., & Nee, B. (2006). San Francisco City CarShare: Longer-term travel-demand and car
ownership impacts (No. 2006, 07). Working Paper.
Cervero, R., Golub, A., & Nee, B. (2007). City CarShare: longer-term travel demand and car ownership
impacts. Transportation Research Record, 1992(1), 70-80.
Che, M., Lum, K. M., & Wong, Y. D. (2020). Users' attitudes on electric scooter riding speed on shared footpath:
A virtual reality study. International Journal of Sustainable Transportation, 1-10.
Chen, M., Wang, D., Sun, Y., Waygood, E. O. D., & Yang, W. (2020). A comparison of users’ characteristics
between station-based bikesharing system and free-floating bikesharing system: Case study in Hangzhou,
China. Transportation, 47(2), 689-704.
Chen, Z., van Lierop, D., & Ettema, D. (2020). Dockless bike-sharing systems: what are the
implications?. Transport Reviews, 40(3), 333-353.
Chicco, A., Diana, M., Rodenbach, J., Matthijs, J., Nehrke, G., Ziesak, M., Horvat, M. (2020). STARS shared
mobility opportunities and challenges for European cities: Deliverable D5.1 - Mobility scenarios of car-
sharing: gap analysis and impacts in the cities of tomorrow.
Choron, R. L., & Sakran, J. V. (2019). The integration of electric scooters: useful technology or public health
problem?. American journal of public health, 109(4), 555-556.
Chow, J. Y., & Sayarshad, H. R. (2014). Symbiotic network design strategies in the presence of coexisting
transportation networks. Transportation Research Part B: Methodological, 62, 13-34.
Ciari, F., & Axhausen, K. W. (2012). Choosing carpooling or car sharing as a mode: Swiss stated choice
experiments. In 91st Annual Meeting of the Transportation Research Board (TRB 2012) (pp. 12-4205).
Transportation Research Board (TRB).
Ciari, F., Balac, M., & Balmer, M. (2015). Modelling the effect of different pricing schemes on free-floating car-
sharing travel demand: a test case for Zurich, Switzerland. Transportation, 42(3), 413-433.

258
Ciari, F., Bock, B., & Balmer, M. (2014). Modeling station-based and free-floating carsharing demand: test case
study for Berlin. Transportation Research Record, 2416(1), 37-47.
Ciociola, A., Cocca, M., Giordano, D., Vassio, L., & Mellia, M. (2020, September). E-Scooter Sharing:
Leveraging Open Data for System Design. In 2020 IEEE/ACM 24th International Symposium on
Distributed Simulation and Real Time Applications (DS-RT) (pp. 1-8). IEEE.
Ciuffini, M., Asperti, S., Gentili, V., Orsini, R., & Refrigeri, L. (2021). 5° rapporto nazionale sulla sharing
mobility. Fondazione per lo Sviluppo Sostenibile, Roma.
Ciuffini, M., Asperti, S., Gentili, V., Orsini, R., & Refrigeri, L. (2022). 6° rapporto nazionale sulla sharing
mobility. Fondazione per lo Sviluppo Sostenibile, Roma.
Clark, M., Gifford, K., Anable, J., & Le Vine, S. (2015). Business-to-business carsharing: evidence from Britain
of factors associated with employer-based carsharing membership and its impacts. Transportation, 42(3),
471-495.
Clewlow, R. R. (2016). Carsharing and sustainable travel behavior: Results from the San Francisco Bay
Area. Transport Policy, 51, 158-164.
Clewlow, R. R. (2019). The Micro-Mobility Revolution: The Introduction and Adoption of Electric Scooters in
the United States (No. 19-03991).
Clifton, K. J. (2003). Independent mobility among teenagers: exploration of travel to after-school
activities. Transportation Research Record, 1854(1), 74-80.
Cobuloglu, H. I., & Büyüktahtakın, İ. E. (2015). A stochastic multi-criteria decision analysis for sustainable
biomass crop selection. Expert Systems with Applications, 42(15-16), 6065-6074.
Fishburn, P. C. (1967). Letter to the editor—additive utilities with incomplete product sets: application to priorities
and assignments. Operations Research, 15(3), 537-542.
Coll, M. H., Vandersmissen, M. H., & Thériault, M. (2014). Modeling spatio-temporal diffusion of carsharing
membership in Québec City. Journal of Transport Geography, 38, 22-37.
ComuneTorino,accessed8december2021, <http://www.comune.torino.it/torinogiovani/vivere-a-torino/sharing-
di-monopattini-elettrici-a-torino>.
Cooper, G., Howe, D. A., & Mye, P. (2000). The Missing Link: An Evaluation of CarSharing Portland Inc.
Portland, Oregon.
Corcoran, J., Li, T., Rohde, D., Charles-Edwards, E., & Mateo-Babiano, D. (2014). Spatio-temporal patterns of a
Public Bicycle Sharing Program: the effect of weather and calendar events. Journal of Transport
Geography, 41, 292-305.
Costain, C., Ardron, C., & Habib, K. N. (2012). Synopsis of users’ behaviour of a carsharing program: A case
study in Toronto. Transportation Research Part A: Policy and Practice, 46(3), 421-434.
Croci, E., & Rossi, D. (2014). Optimizing the position of bike sharing stations. The Milan case.
Curl, A., & Fitt, H. (2020). Same same, but different? Cycling and e‐scootering in a rapidly changing urban
transport landscape. New Zealand Geographer, 76(3), 194-206.

259
Davis, B., Dutzik, T., & Baxandall, P. (2012). Transportation and the new generation: Why young people are
driving less and what it means for transportation policy.
De Lorimier, A., & El-Geneidy, A. M. (2013). Understanding the factors affecting vehicle usage and availability
in carsharing networks: A case study of Communauto carsharing system from Montréal,
Canada. International Journal of Sustainable Transportation, 7(1), 35-51.
De Luca, S., & Di Pace, R. (2015). Modelling users’ behaviour in inter-urban carsharing program: A stated
preference approach. Transportation research part A: policy and practice, 71, 59-76.
Del Mar Alonso-Almeida, M. (2019). Carsharing: Another gender issue? Drivers of carsharing usage among
women and relationship to perceived value. Travel behaviour and society, 17, 36-45.
Dell’Olio, L., Ibeas, A., & Cecín, P. (2010). Modelling user perception of bus transit quality. Transport
Policy, 17(6), 388-397.
Dell’Olio, L., Ibeas, A., Bordagaray, M., & Ortúzar, J. D. D. (2014). Modeling the effects of pro bicycle
infrastructure and policies toward sustainable urban mobility. Journal of Urban Planning and
Development, 140(2), 04014001.
Dell'Olio, L., Ibeas, A., & Moura, J. L. (2011, June). Implementing bike-sharing systems. In Proceedings of the
Institution of Civil Engineers-Municipal Engineer (Vol. 164, No. 2, pp. 89-101). Thomas Telford Ltd.
DeMaio, P. (2009). Bike-sharing: History, impacts, models of provision, and future. Journal of public
transportation, 12(4), 3.
DeMaio, P. J. (2003). Smart bikes: Public transportation for the 21st century. Transportation Quarterly, 57(1), 9-
11.
Dias, F. F., Lavieri, P. S., Garikapati, V. M., Astroza, S., Pendyala, R. M., & Bhat, C. R. (2017). A behavioral
choice model of the use of car-sharing and ride-sourcing services. Transportation, 44(6), 1307-1323.
Dill, J. (2009). Bicycling for transportation and health: the role of infrastructure. Journal of public health
policy, 30(1), S95-S110.
Dill, J., & Carr, T. (2003). Bicycle commuting and facilities in major US cities: if you build them, commuters will
use them. Transportation research record, 1828(1), 116-123.
Dill, J., McNeil, N., & Howland, S. (2019). Effects of peer-to-peer carsharing on vehicle owners’ travel
behavior. Transportation Research Part C: Emerging Technologies, 101, 70-78.
Dörry, S., & Decoville, A. (2016). Governance and transportation policy networks in the cross-border
metropolitan region of Luxembourg: A social network analysis. European Urban and Regional
Studies, 23(1), 69-85.
Douma, F., Gaug, R., Horan, T., & Schooley, B. (2008). Improving Carsharing and Transit Service with ITS.
Du, M., & Cheng, L. (2018). Better understanding the characteristics and influential factors of different travel
patterns in free-floating bike sharing: Evidence from Nanjing, China. Sustainability, 10(4), 1244.
Du, M., Cheng, L., Li, X., & Yang, J. (2019). Investigating the Influential Factors of Shared Travel Behavior:
Comparison between App-Based Third Taxi Service and Free-Floating Bike Sharing in Nanjing,
China. Sustainability, 11(16), 4318.

260
Duan, Q., Ye, X., Li, J., & Wang, K. (2020). Empirical modeling analysis of potential commute demand for
carsharing in shanghai, China. Sustainability, 12(2), 620.
Duran-Rodas, D., Chaniotakis, E., & Antoniou, C. (2019). Built Environment Factors Affecting Bike Sharing
Ridership: Data-Driven Approach for Multiple Cities. Transportation Research Record,
0361198119849908.
Efthymiou, D., & Antoniou, C. (2014). Modeling the propensity to join carsharing using hybrid choice and latent
variable models 2 and mixed internet/paper survey data 3 (No. 14-2512).
Efthymiou, D., & Antoniou, C. (2016). Modeling the propensity to join carsharing using hybrid choice models
and mixed survey data. Transport Policy, 51, 143-149.
Efthymiou, D., Antoniou, C., & Waddell, P. (2013). Factors affecting the adoption of vehicle sharing systems by
young drivers. Transport policy, 29, 64-73.
El-Assi, W., Mahmoud, M. S., & Habib, K. N. (2017). Effects of built environment and weather on bike sharing
demand: a station level analysis of commercial bike sharing in Toronto. Transportation, 44(3), 589-613.
Elhenawy, M., Komol, M. D., Ashqar, H. I., Almannaa, M. H., Masoud, M., Rakha, H. A., & Rakotonirainy, A.
(2020). Developing a novel crowdsourcing business model for micro-mobility ride-sharing systems:
Methodology and preliminary results. arXiv preprint arXiv:2007.15585.
Eren, E., & Uz, V. E. (2020). A review on bike-sharing: The factors affecting bike-sharing demand. Sustainable
Cities and Society, 54, 101882.
Etienne, C., & Latifa, O. (2014). Model-based count series clustering for bike sharing system usage mining: a
case study with the Vélib’system of Paris. ACM Transactions on Intelligent Systems and Technology
(TIST), 5(3), 1-21.
Faghih-Imani, A., & Eluru, N. (2015). Analyzing bicycle-sharing system user destination choice preferences:
Chicago’s Divvy system. Journal of transport geography, 44, 53-64.
Faghih-Imani, A., & Eluru, N. (2016a). Determining the role of bicycle sharing system infrastructure installation
decision on usage: Case study of montreal BIXI system. Transportation Research Part A: Policy and
Practice, 94, 685-698.
Faghih-Imani, A., & Eluru, N. (2016b). Incorporating the impact of spatio-temporal interactions on bicycle
sharing system demand: A case study of New York CitiBike system. Journal of Transport Geography, 54,
218-227.
Faghih-Imani, A., Eluru, N., El-Geneidy, A. M., Rabbat, M., & Haq, U. (2014). How land-use and urban form
impact bicycle flows: evidence from the bicycle-sharing system (BIXI) in Montreal. Journal of Transport
Geography, 41, 306-314.
Faghih-Imani, A., Hampshire, R., Marla, L., & Eluru, N. (2017). An empirical analysis of bike sharing usage and
rebalancing: Evidence from Barcelona and Seville. Transportation Research Part A: Policy and
Practice, 97, 177-191.
Fang, K., Agrawal, A. W., Steele, J., Hunter, J. J., & Hooper, A. M. (2018). Where Do Riders Park Dockless,
Shared Electric Scooters? Findings from San Jose, California.
Fawcett, C. R., Barboza, D., Gasvoda, H. L., & Bernier, M. D. (2018). Analyzing Rideshare Bicycles and Scooters.

261
Feng, C., Jiao, J., & Wang, H. (2020). Estimating E-Scooter Traffic Flow Using Big Data to Support Planning for
Micromobility. Journal of Urban Technology, 1-19.
Feng, P., & Li, W. (2016). Willingness to use a public bicycle system: An example in Nanjing City. Journal of
Public Transportation, 19(1), 6.
Ferrero, F., Perboli, G., Rosano, M., & Vesco, A. (2018). Car-sharing services: An annotated review. Sustainable
Cities and Society, 37, 501-518.
Firnkorn, J. (2012). Triangulation of two methods measuring the impacts of a free-floating carsharing system in
Germany. Transportation Research Part A: Policy and Practice, 46(10), 1654-1672.
Firnkorn, J., & Müller, M. (2011). What will be the environmental effects of new free-floating car-sharing
systems? The case of car2go in Ulm. Ecological economics, 70(8), 1519-1528.
Firnkorn, J., & Müller, M. (2012). Selling mobility instead of cars: new business strategies of automakers and the
impact on private vehicle holding. Business Strategy and the environment, 21(4), 264-280.
Firnkorn, J., & Shaheen, S. (2016). Generic time-and method-interdependencies of empirical impact-
measurements: A generalizable model of adaptation-processes of carsharing-users' mobility-behavior
over time. Journal of Cleaner Production, 113, 897-909.
Fishman, E. (2016). Bikeshare: A review of recent literature. Transport Reviews, 36(1), 92-113.
Fishman, E., Washington, S., & Haworth, N. (2013). Bike share: a synthesis of the literature. Transport
reviews, 33(2), 148-165.
Fishman, E., Washington, S., Haworth, N., & Mazzei, A. (2014). Barriers to bikesharing: an analysis from
Melbourne and Brisbane. Journal of Transport Geography, 41, 325-337.
Fishman, E., Washington, S., Haworth, N., & Watson, A. (2015). Factors influencing bike share membership: An
analysis of Melbourne and Brisbane. Transportation research part A: policy and practice, 71, 17-30.
Fishman, L., & Wabe, J. S. (1968). Restructuring the form of car ownership (No. 2068-2018-951).
Fleury, S., Tom, A., Jamet, E., & Colas-Maheux, E. (2017). What drives corporate carsharing acceptance? A
French case study. Transportation Research Part F: Traffic Psychology and Behaviour, 45, 218-227.
Flügel, S., Ramjerdi, F., Veisten, K., Killi, M., & Elvik, R. (2015). Valuation of cycling facilities with and without
controlling for casualty risk. International journal of sustainable transportation, 9(5), 364-376.
Forman, E. H. (1990). Random indices for incomplete pairwise comparison matrices. European journal of
operational research, 48(1), 153-155.
Forman, E., & Peniwati, K. (1998). Aggregating individual judgments and priorities with the analytic hierarchy
process. European journal of operational research, 108(1), 165-169.
Forsyth, A., & Oakes, J. M. (2015). Cycling, the built environment, and health: results of a midwestern
study. International Journal of Sustainable Transportation, 9(1), 49-58.
Fournier, N., Christofa, E., & Knodler Jr, M. A. (2017). A sinusoidal model for seasonal bicycle demand
estimation. Transportation research part D: transport and environment, 50, 154-169.

262
Frade, I., & Ribeiro, A. (2014). Bicycle sharing systems demand. Procedia-Social and Behavioral Sciences, 111,
518-527.
Fricker, C., & Gast, N. (2016). Incentives and redistribution in homogeneous bike-sharing systems with stations
of finite capacity. Euro journal on transportation and logistics, 5(3), 261-291
Froehlich, J., Neumann, J., & Oliver, N. (2008). Measuring the pulse of the city through shared bicycle
programs. Proc. of UrbanSense08, 16-20.
Fuller, D., Gauvin, L., Kestens, Y., Daniel, M., Fournier, M., Morency, P., & Drouin, L. (2011). Use of a new
public bicycle share program in Montreal, Canada. American journal of preventive medicine, 41(1), 80-
83.
Fuller, D., Gauvin, L., Kestens, Y., Daniel, M., Fournier, M., Morency, P., & Drouin, L. (2013). Impact evaluation
of a public bicycle share program on cycling: a case example of BIXI in Montreal, Quebec. American
journal of public health, 103(3), e85-e92.
Garrard, J., Rose, G., & Lo, S. K. (2008). Promoting transportation cycling for women: the role of bicycle
infrastructure. Preventive medicine, 46(1), 55-59.
Gebhart, K., & Noland, R. B. (2014). The impact of weather conditions on bikeshare trips in Washington,
DC. Transportation, 41(6), 1205-1225.
Geldermann, J., & Rentz, O. (2001). Integrated technique assessment with imprecise information as a support for
the identification of best available techniques (BAT). OR-Spektrum, 23(1), 137-157.
Gibson, H. (2020). Contested boundaries: E-scooter riders’ and pedestrians’ experiences of sharing
space (Doctoral dissertation, University of Otago).
Gilks, W. R., Richardson, S., & Spiegelhalter, D. (Eds.). (1995). Markov chain Monte Carlo in practice. CRC
press.
Glassman, R. B., Garvey, K. J., Elkins, K. M., Kasal, K. L., & Couillard, N. L. (1994). Spatial working memory
score of humans in a large radial maze, similar to published score of rats, implies capacity close to the
magical number 7±2. Brain research bulletin, 34(2), 151-159.
Gleason, R., & Miskimins, L. (2012). Exploring bicycle options for federal lands: bike sharing, rentals and
employee fleets (No. FHWA-WFL/TD-12-001).
Glotz-Richter, M. (2016). Reclaim street space!–exploit the European potential of car sharing. Transportation
Research Procedia, 14, 1296-1304.
Godavarthy, R. P., & Taleqani, A. R. (2017). Winter bikesharing in US: User willingness, and operator’s
challenges and best practices. Sustainable cities and society, 30, 254-262.
Golden, B. L., & Wang, Q. (1989). An alternate measure of consistency. In The analytic hierarchy process (pp.
68-81). Springer, Berlin, Heidelberg.
Goodman, A., & Cheshire, J. (2014). Inequalities in the London bicycle sharing system revisited: impacts of
extending the scheme to poorer areas but then doubling prices. Journal of Transport Geography, 41, 272-
279.
Gössling, S. (2020). Integrating e-scooters in urban transportation: Problems, policies, and the prospect of system
change. Transportation Research Part D: Transport and Environment, 79, 102230.

263
Grenier, T., Deckelbaum, D. L., Boulva, K., Drudi, L., Feyz, M., Rodrigue, N., ... & Razek, T. (2013). A
descriptive study of bicycle helmet use in Montreal, 2011. Canadian journal of public health, 104(5),
e400-e404.
Groenendijk, L., Rezaei, J., & Correia, G. (2018). Incorporating the travellers’ experience value in assessing the
quality of transit nodes: A Rotterdam case study. Case Studies on Transport Policy, 6(4), 564-576.
Guirao, B., Ampudia, M., Molina, R., & García-Valdecasas, J. (2018). Student behaviour towards Free-Floating
Carsharing: First evidences of the experience in Madrid. Transportation research procedia, 33, 243-250.
Gupta, H. (2018). Evaluating service quality of airline industry using hybrid best worst method and
VIKOR. Journal of Air Transport Management, 68, 35-47.
Gupta, H., & Barua, M. K. (2016). Identifying enablers of technological innovation for Indian MSMEs using
best–worst multi criteria decision making method. Technological Forecasting and Social Change, 107,
69-79.
Gupta, H., & Barua, M. K. (2018). A novel hybrid multi-criteria method for supplier selection among SMEs on
the basis of innovation ability. International Journal of Logistics Research and Applications, 21(3), 201-
223.
Habib, K. M. N., Morency, C., Islam, M. T., & Grasset, V. (2012). Modelling users’ behaviour of a carsharing
program: Application of a joint hazard and zero inflated dynamic ordered probability
model. Transportation research part A: policy and practice, 46(2), 241-254.
Hafezalkotob, A., & Hafezalkotob, A. (2017). A novel approach for combination of individual and group
decisions based on fuzzy best-worst method. Applied Soft Computing, 59, 316-325.
Hafezalkotob, A., Hafezalkotob, A., Liao, H., & Herrera, F. (2019). Interval MULTIMOORA method integrating
interval Borda rule and interval best–worst-method-based weighting model: case study on hybrid vehicle
engine selection. IEEE transactions on cybernetics, 50(3), 1157-1169.
Hafezalkotob, A., Hami-Dindar, A., Rabie, N., & Hafezalkotob, A. (2018). A decision support system for
agricultural machines and equipment selection: A case study on olive harvester machines. Computers
and Electronics in Agriculture, 148, 207-216.
Hampshire, R. C., & Marla, L. (2012, January). An analysis of bike sharing usage: Explaining trip generation and
attraction from observed demand. In 91st Annual meeting of the transportation research board,
Washington, DC (pp. 12-2099).
Handy, S. L., Xing, Y., & Buehler, T. J. (2010). Factors associated with bicycle ownership and use: a study of six
small US cities. Transportation, 37(6), 967-985.
Haworth, N. L., & Schramm, A. (2019). Illegal and risky riding of electric scooters in Brisbane. Medical journal
of Australia, 211(9), 412-413.
He, S., & Shin, K. G. (2020, April). Dynamic Flow Distribution Prediction for Urban Dockless E-Scooter Sharing
Reconfiguration. In Proceedings of The Web Conference 2020 (pp. 133-143).
Heaney, A. K., Carrión, D., Burkart, K., Lesk, C., & Jack, D. (2019). Climate Change and Physical Activity:
Estimated Impacts of Ambient Temperatures on Bikeshare Usage in New York City. Environmental
health perspectives, 127(3), 037002.

264
Hollingsworth, J., Copeland, B., & Johnson, J. X. (2019). Are e-scooters polluters? The environmental impacts
of shared dockless electric scooters. Environmental Research Letters, 14(8), 084031.
Hu, S., Chen, P., Lin, H., Xie, C., & Chen, X. (2018). Promoting carsharing attractiveness and efficiency: An
exploratory analysis. Transportation Research Part D: Transport and Environment, 65, 229-243.
Hua, Y., Zhao, D., Wang, X., & Li, X. (2019). Joint infrastructure planning and fleet management for one-way
electric car sharing under time-varying uncertain demand. Transportation Research Part B:
Methodological, 128, 185-206.
Hunt, J. D., & Abraham, J. E. (2007). Influences on bicycle use. Transportation, 34(4), 453-470.
Huwer, U. (2004). Public transport and csar-sharing—benefits and effects of combined services. Transport
Policy, 11(1), 77-87.
Hwang, C. L., & Yoon, K. (1981). Multiple Attribute Decision Making–Methods and Applications Springer-
Verlag Berlin Heidelberg. New York.
Hyland, M., Hong, Z., de Farias Pinto, H. K. R., & Chen, Y. (2018). Hybrid cluster-regression approach to model
bikeshare station usage. Transportation Research Part A: Policy and Practice, 115, 71-89.
Jain, T., Wang, X., Rose, G., & Johnson, M. (2018). Does the role of a bicycle share system in a city change over
time? A longitudinal analysis of casual users and long-term subscribers. Journal of transport
geography, 71, 45-57.
James, O., Swiderski, J. I., Hicks, J., Teoman, D., & Buehler, R. (2019). Pedestrians and e-scooters: An initial
look at e-scooter parking and perceptions by riders and non-riders. Sustainability, 11(20), 5591.
Jäppinen, S., Toivonen, T., & Salonen, M. (2013). Modelling the potential effect of shared bicycles on public
transport travel times in Greater Helsinki: An open data approach. Applied Geography, 43, 13-24.
Jara-Díaz, S. R., & Videla, J. (1989). Detection of income effect in mode choice: theory and
application. Transportation Research Part B: Methodological, 23(6), 393-400.
Jennings, G. (2011). A challenge shared: is South African ready for a public bicycle system?. SATC 2011.
Jensen, P., Rouquier, J. B., Ovtracht, N., & Robardet, C. (2010). Characterizing the speed and paths of shared
bicycle use in Lyon. Transportation research part D: transport and environment, 15(8), 522-524.
Ji, Y., Ma, X., He, M., Jin, Y., & Yuan, Y. (2020). Comparison of usage regularity and its determinants between
docked and dockless bike-sharing systems: A case study in Nanjing, China. Journal of Cleaner
Production, 255, 120110.
Jia, L., Liu, X., & Liu, Y. (2018). Impact of different stakeholders of bike-sharing industry on users’ intention of
civilized use of bike-sharing. Sustainability, 10(5), 1437.
Jian, S., Rashidi, T. H., & Dixit, V. (2017). An analysis of carsharing vehicle choice and utilization patterns using
multiple discrete-continuous extreme value (MDCEV) models. Transportation Research Part A: Policy
and Practice, 103, 362-376.
Jin, F., An, K., & Yao, E. (2020). Mode choice analysis in urban transport with shared battery electric vehicles:
A stated-preference case study in Beijing, China. Transportation Research Part A: Policy and
Practice, 133, 95-108.

265
Jing, C., & Zhao, Z. (2015, July). Research on Antecedents and Consequences of Factors Affecting the Bike
Sharing System---Lessons From Capital Bike Share Program in Washington, DC. In International
Conference on Logistics Engineering, Management and Computer Science (LEMCS 2015). Atlantis
Press.
Jones, E. C., & Leibowicz, B. D. (2019). Contributions of shared autonomous vehicles to climate change
mitigation. Transportation Research Part D: Transport and Environment, 72, 279-298.
Jong, N. K., & Stone, P. (1976). Keeney, RL &Raiffa, H. Decisions with Multiple Objectives: Preferences and
Value Tradeoffs. In In Proceedings of the ICML-06 Workshop on Kernel Methods in Reinforcement
Learning.
Jorge, D., & Correia, G. (2013). Carsharing systems demand estimation and defined operations: a literature
review. European Journal of Transport and Infrastructure Research, 13(3).
Jorge, D., Barnhart, C., & de Almeida Correia, G. H. (2015). Assessing the viability of enabling a round-trip
carsharing system to accept one-way trips: Application to Logan Airport in Boston. Transportation
Research Part C: Emerging Technologies, 56, 359-372.
Jorge, D., Molnar, G., & de Almeida Correia, G. H. (2015). Trip pricing of one-way station-based car-sharing
networks with zone and time of day price variations. Transportation Research Part B:
Methodological, 81, 461-482.
Juschten, M., Ohnmacht, T., Thao, V. T., Gerike, R., & Hössinger, R. (2017). Carsharing in Switzerland:
identifying new markets by predicting membership based on data on supply and
demand. Transportation, 46(4), 1171-1194.
Kabak, M., Erbaş, M., Çetinkaya, C., & Özceylan, E. (2018). A GIS-based MCDM approach for the evaluation
of bike-share stations. Journal of cleaner production, 201, 49-60.
Kalpoe, R. (2020a). A multi-criteria assessment to determine the customers’ technology preference in the context
of apparel e-commerce.
Kalpoe, R. (2020b). Technology acceptance and return management in apparel e-commerce. Journal of Supply
Chain Management Science, 1(3-4).
Kaltenbrunner, A., Meza, R., Grivolla, J., Codina, J., & Banchs, R. (2010). Urban cycles and mobility patterns:
Exploring and predicting trends in a bicycle-based public transport system. Pervasive and Mobile
Computing, 6(4), 455-466.
Kamargianni, M. (2015). Investigating next generation's cycling ridership to promote sustainable mobility in
different types of cities. Research in Transportation Economics, 53, 45-55.
Kamargianni, M., & Polydoropoulou, A. (2013). Hybrid choice model to investigate effects of teenagers' attitudes
toward walking and cycling on mode choice behavior. Transportation research record, 2382(1), 151-161.
Karimi, H., Sadeghi-Dastaki, M., & Javan, M. (2020). A fully fuzzy best–worst multi attribute decision making
method with triangular fuzzy number: A case study of maintenance assessment in the hospitals. Applied
Soft Computing, 86, 105882.
Kaspi, M., Raviv, T., & Tzur, M. (2014). Parking reservation policies in one-way vehicle sharing
systems. Transportation Research Part B: Methodological, 62, 35-50.

266
Katona, G., & Juhasz, J. (2020). The History of the Transport System Development and Future with Sharing and
Autonomous Systems. Communications-Scientific letters of the University of Zilina, 22(1), 25-34.
Kawgan-Kagan, I. (2015). Early adopters of carsharing with and without BEVs with respect to gender
preferences. European Transport Research Review, 7(4), 33.
Kazmaier, M., Taefi, T. T., & Hettesheimer, T. (2020). Techno-economical and ecological potential of electric
scooters: a life cycle analysis. European Journal of Transport and Infrastructure Research, 20(4), 233-
251.
Kek, A. G., Cheu, R. L., & Chor, M. L. (2006). Relocation simulation model for multiple-station shared-use
vehicle systems. Transportation research record, 1986(1), 81-88.
Kent, J. L., & Dowling, R. (2016). The future of paratransit and DRT: Introducing cars on demand. In Paratransit:
Shaping the Flexible Transport Future. Emerald Group Publishing Limited.
Kerr, J., Rosenberg, D., Sallis, J. F., Saelens, B. E., Frank, L. D., & Conway, T. L. (2006). Active commuting to
school: associations with environment and parental concerns. Medicine & Science in Sports &
Exercise, 38(4), 787-793.
Khandelwal, M. (2021). https://www.surveysensum.com/blog/everything-you-need-to-know-about-the-likert-
scale/- accessed on 10, January, 2022.
Kim, D., Ko, J., & Park, Y. (2015). Factors affecting electric vehicle sharing program participants’ attitudes about
car ownership and program participation. Transportation Research Part D: Transport and
Environment, 36, 96-106.
Kim, D., Shin, H., Im, H., & Park, J. (2012). Factors influencing travel behaviors in bikesharing. In Transportation
research board 91st annual meeting. Washington, DC: Transportation Research Board (pp. 1-14).
Kim, J., Rasouli, S., & Timmermans, H. J. (2017a). Investigating heterogeneity in social influence by social
distance in car-sharing decisions under uncertainty: A regret-minimizing hybrid choice model framework
based on sequential stated adaptation experiments. Transportation Research Part C: Emerging
Technologies, 85, 47-63.
Kim, J., Rasouli, S., & Timmermans, H. J. (2017b). Satisfaction and uncertainty in car-sharing decisions: An
integration of hybrid choice and random regret-based models. Transportation Research Part A: Policy
and Practice, 95, 13-33.
Kim, J., Rasouli, S., & Timmermans, H. J. (2017c). The effects of activity-travel context and individual attitudes
on car-sharing decisions under travel time uncertainty: A hybrid choice modeling
approach. Transportation Research Part D: Transport and Environment, 56, 189-202.
Kim, K. (2018). Investigation on the effects of weather and calendar events on bike-sharing according to the trip
patterns of bike rentals of stations. Journal of transport geography, 66, 309-320.
Kolios, A., Mytilinou, V., Lozano-Minguez, E., & Salonitis, K. (2016). A comparative study of multiple-criteria
decision-making methods under stochastic inputs. Energies, 9(7), 566.
Kopp, J., Gerike, R., & Axhausen, K. W. (2015). Do sharing people behave differently? An empirical evaluation
of the distinctive mobility patterns of free-floating car-sharing members. Transportation, 42(3), 449-469.
Kortum, K., & Machemehl, R. B. (2012). Free-floating carsharing systems: innovations in membership prediction,
mode share, and vehicle allocation optimization methodologies (No. SWUTC/12/476660-00079-1).

267
Southwest Region University Transportation Center, Center for Transportation Research, University of
Texas at Austin.
Kraemer, J. D., Roffenbender, J. S., & Anderko, L. (2012). Helmet wearing among users of a public bicycle-
sharing program in the District of Columbia and comparable riders on personal bicycles. American
journal of public health, 102(8), e23-e25.
Krizek, K. J., & Roland, R. W. (2005). What is at the end of the road? Understanding discontinuities of on-street
bicycle lanes in urban settings. Transportation Research Part D: Transport and Environment, 10(1), 55-
68.
Krizek, K. J., El-Geneidy, A., & Thompson, K. (2007). A detailed analysis of how an urban trail system affects
cyclists’ travel. Transportation, 34(5), 611-624.
Kruschke, J. (2014). Doing Bayesian data analysis: A tutorial with R, JAGS, and Stan.
Krykewycz, G. R., Puchalsky, C. M., Rocks, J., Bonnette, B., & Jaskiewicz, F. (2010). Defining a primary market
and estimating demand for major bicycle-sharing program in Philadelphia, Pennsylvania. Transportation
Research Record, 2143(1), 117-124.
Kusi-Sarpong, S., Gupta, H., & Sarkis, J. (2019). A supply chain sustainability innovation framework and
evaluation methodology. International Journal of Production Research, 57(7), 1990-2008.
Kutela, B., & Kidando, E. (2017). Towards a Better Understanding of Effectiveness of Bike-share Programs:
Exploring Factors Affecting Bikes Idle Duration. American Scientific Research Journal for Engineering,
Technology, and Sciences (ASRJETS), 29(1), 33-46.
Laa, B., & Leth, U. (2020). Survey of E-scooter users in Vienna: Who they are and how they ride. Journal of
transport geography, 89, 102874.
Lachance-Bernard, N., Produit, T., Tominc, B., Nikšič, M., & Marušić, B. G. (2011, June). Network based kernel
density estimation for cycling facilities optimal location applied to Ljubljana. In International
Conference on Computational Science and Its Applications (pp. 136-150). Springer, Berlin, Heidelberg.
Lagadic, M., Verloes, A., & Louvet, N. (2019). Can carsharing services be profitable? A critical review of
established and developing business models. Transport Policy, 77(C), 68-78.
Lamberton, C. P., & Rose, R. L. (2012). When is ours better than mine? A framework for understanding and
altering participation in commercial sharing systems. Journal of Marketing, 76(4), 109-125.
Lan, J., Ma, Y., Zhu, D., Mangalagiu, D., & Thornton, T. F. (2017). Enabling value co-creation in the sharing
economy: The case of mobike. Sustainability, 9(9), 1504.
Lane, C. (2005). PhillyCarShare: First-year social and mobility impacts of carsharing in Philadelphia,
Pennsylvania. Transportation Research Record, 1927(1), 158-166.
Lane, E. F., & Verdini, W. A. (1989). A consistency test for AHP decision makers. Decision Sciences, 20(3), 575-
590.
Larsen, J., & El-Geneidy, A. (2011). A travel behavior analysis of urban cycling facilities in Montréal,
Canada. Transportation research part D: transport and environment, 16(2), 172-177.

268
Larsen, K., Gilliland, J., Hess, P., Tucker, P., Irwin, J., & He, M. (2009). The influence of the physical environment
and sociodemographic characteristics on children's mode of travel to and from school. American Journal
of Public Health, 99(3), 520-526.
Le Pira, M., Ignaccolo, M., Inturri, G., Pluchino, A., & Rapisarda, A. (2016). Modelling stakeholder participation
in transport planning. Case Studies on Transport Policy, 4(3), 230-238.
Le Pira, M., Marcucci, E., Gatta, V., Ignaccolo, M., Inturri, G., & Pluchino, A. (2017). Towards a decision-support
procedure to foster stakeholder involvement and acceptability of urban freight transport
policies. European Transport Research Review, 9(4), 1-14.
Le Vine, S., & Polak, J. (2019). The impact of free-floating carsharing on car ownership: Early-stage findings
from London. Transport Policy, 75, 119-127.
Le Vine, S., Adamou, O., & Polak, J. (2014). Predicting new forms of activity/mobility patterns enabled by shared-
mobility services through a needs-based stated-response method: Case study of grocery
shopping. Transport Policy, 32, 60-68.
Le Vine, S., Lee-Gosselin, M., Sivakumar, A., & Polak, J. (2014). A new approach to predict the market and
impacts of round-trip and point-to-point carsharing systems: case study of London. Transportation
Research Part D: Transport and Environment, 32, 218-229.
Lee, M., Chow, J. Y., Yoon, G., & He, B. Y. (2019). Forecasting e-scooter competition with direct and access
trips by mode and distance in New York City. arXiv preprint arXiv:1908.08127.
Leister, E. H., Vairo, N., Sims, D., & Bopp, M. (2018). Understanding bike share reach, use, access and function:
An exploratory study. Sustainable cities and society, 43, 191-196.
Lempert, R., Zhao, J., & Dowlatabadi, H. (2019). Convenience, savings, or lifestyle? Distinct motivations and
travel patterns of one-way and two-way carsharing members in Vancouver, Canada. Transportation
Research Part D: Transport and Environment, 71, 141-152.
Levy, N., Golani, C., & Ben-Elia, E. (2019). An exploratory study of spatial patterns of cycling in Tel Aviv using
passively generated bike-sharing data. Journal of Transport Geography, 76, 325-334.
Li, L., Lee, K. Y., Yang, S. B., & Chang, L. Y. (2020, January). Linking Privacy Concerns for Traceable
Information and Information Privacy Protective Responses on Electric Scooter Sharing Platforms.
In Proceedings of the 53rd Hawaii International Conference on System Sciences.
Li, L., Liu, Y., & Song, Y. (2019). Factors affecting bike-sharing behaviour in Beijing: price, traffic congestion,
and supply chain. Annals of Operations Research, 1-16.
Li, L., Wang, X., & Rezaei, J. (2020). A Bayesian best-worst method-based multicriteria competence analysis of
crowdsourcing delivery personnel. Complexity, 2020.
Li, Q., Liao, F., Timmermans, H. J., Huang, H., & Zhou, J. (2018). Incorporating free-floating car-sharing into an
activity-based dynamic user equilibrium model: A demand-side model. Transportation Research Part B:
Methodological, 107, 102-123.
Li, W. (2019). A mode choice study on shared mobility services: Policy opportunities for a developing
country (Doctoral dissertation, UCL (University College London)).

269
Li, W., & Kamargianni, M. (2018). Providing quantified evidence to policy makers for promoting bike-sharing in
heavily air-polluted cities: A mode choice model and policy simulation for Taiyuan-
China. Transportation research part A: policy and practice, 111, 277-291.
Li, X., Zhang, Y., Du, M., & Yang, J. (2019). Social factors influencing the choice of bicycle: Difference analysis
among private bike, public bike sharing and free-floating bike sharing in Kunming, China. KSCE Journal
of Civil Engineering, 23(5), 2339-2348.
Li, X., Zhang, Y., Sun, L., & Liu, Q. (2018). Free-floating bike sharing in jiangsu: Users’ behaviors and
influencing factors. Energies, 11(7), 1664.
Liang, F., Brunelli, M., & Rezaei, J. (2020). Consistency issues in the best worst method: Measurements and
thresholds. Omega, 96, 102175.
Lin, J. R., & Yang, T. H. (2011). Strategic design of public bicycle sharing systems with service level
constraints. Transportation research part E: logistics and transportation review, 47(2), 284-294.
Lin, P., Weng, J., Liang, Q., Alivanistos, D., & Ma, S. (2020). Impact of weather conditions and built environment
on public bikesharing trips in Beijing. Networks and Spatial Economics, 20(1), 1-17.
Ling, Z., Cherry, C. R., Yang, H., & Jones, L. R. (2015). From e-bike to car: A study on factors influencing
motorization of e-bike users across China. Transportation Research Part D: Transport and
Environment, 41, 50-63.
Litman, T. (2000). Evaluating carsharing benefits. Transportation Research Record, 1702(1), 31-35.
Litman, T., & Laube, F. (2002). Automobile dependency and economic development. Victoria Transport Policy
Institute, Canada.
Liu, W. (2016). Determining the Importance of Factors for Transport Modes in Freight Transportation. Master’s
Thesis, Delft University of Technology, Delft, The Netherlands.
Loose, W., Mohr, M., & Nobis, C. (2006). Assessment of the future development of car sharing in Germany and
related opportunities. Transport Reviews, 26(3), 365-382.
Lootsma, F. A. (1990). The French and the American school in multi-criteria decision analysis. RAIRO-
Operations Research-Recherche Opérationnelle, 24(3), 263-285.
Lu, W., Scott, D. M., & Dalumpines, R. (2018). Understanding bike share cyclist route choice using GPS data:
Comparing dominant routes and shortest paths. Journal of transport geography, 71, 172-181.
Ma, Q., Yang, H., Ma, Y., Yang, D., Hu, X., & Xie, K. (2021). Examining municipal guidelines for users of shared
E-Scooters in the United States. Transportation Research Part D: Transport and Environment, 92,
102710.
Ma, Q., Yang, H., Mayhue, A., Sun, Y., Huang, Z., & Ma, Y. (2021). E-scooter safety: the riding risk analysis
based on mobile sensing data. Accident Analysis & Prevention, 151, 105954.
Ma, X., Cao, R., & Wang, J. (2019). Effects of psychological factors on modal shift from car to dockless bike
sharing: a case study of Nanjing, China. International journal of environmental research and public
health, 16(18), 3420.
Macharis, C., 2005. The importance of stakeholder analysis in freight transport: The MAMCA methodology.
European Transport/Transporti Europei 25 (26), 114–120.

270
Macharis, C., De Witte, A., & Turcksin, L. (2010). The Multi-Actor Multi-Criteria Analysis (MAMCA)
application in the Flemish long-term decision making process on mobility and logistics. Transport
Policy, 17(5), 303-311.
Macharis, C., Turcksin, L., & Lebeau, K. (2012). Multi actor multi criteria analysis (MAMCA) as a tool to support
sustainable decisions: State of use. Decision Support Systems, 54(1), 610-620.
Maiti, A., Vinayaga-Sureshkanth, N., Jadliwala, M., Wijewickrama, R., & Griffin, G. P. (2020). Impact of E-
Scooters on Pedestrian Safety: A Field Study Using Pedestrian Crowd-Sensing.
Martin, E., & Shaheen, S. (2011a). The impact of carsharing on public transit and non-motorized travel: an
exploration of North American carsharing survey data. Energies, 4(11), 2094-2114.
Martin, E., & Shaheen, S. (2011b). Greenhouse gas emission impacts of carsharing in North America. IEEE
Transactions on intelligent transportation systems, 12(4), 1074-1086.
Martin, E., Shaheen, S. (2016). Impacts of car2go on Vehicle Ownership, Modal Shift, Vehicle Miles Travelled,
and Greenhouse Gas Emissions: An Analysis of Five North American Cities. Berkeley.
Martin, E., Shaheen, S. A., & Lidicker, J. (2010). Impact of carsharing on household vehicle holdings: Results
from North American shared-use vehicle survey. Transportation Research Record, 2143(1), 150-158.
Martínez, L. M., Correia, G. H. D. A., Moura, F., & Mendes Lopes, M. (2017). Insights into car-sharing demand
dynamics: Outputs of an agent-based model application to Lisbon, Portugal. International Journal of
Sustainable Transportation, 11(2), 148-159.
Martinez, M. (2017). The impact weather has on NYC Citi Bike share company activity. Journal of Environmental
and Resource Economics at Colby, 4(1), 12.
Mateo, J. R. S. C. (2012). Multi-attribute utility theory. In Multi Criteria Analysis in the Renewable Energy
Industry (pp. 63-72). Springer, London.
Mateo-Babiano, I., Bean, R., Corcoran, J., & Pojani, D. (2016). How does our natural and built environment affect
the use of bicycle sharing?. Transportation Research Part A: Policy and Practice, 94, 295-307.
Mathew, J. K., Liu, M., Seeder, S., Li, H., & Bullock, D. M. (2019). Analysis of e-scooter trips and their temporal
usage patterns, Institute Transportation. Engineering Journal, 89, 44–49.
Maurer, L. K. (2011). Suitability study for a bicycle sharing program in Sacramento, California.
Mayhew, L. J., & Bergin, C. (2019). Impact of e‐scooter injuries on Emergency Department imaging. Journal of
medical imaging and radiation oncology, 63(4), 461-466.
McBain, C., & Caulfield, B. (2018). An analysis of the factors influencing journey time variation in the cork
public bike system. Sustainable cities and society, 42, 641-649.
McKenzie, G. (2019). Spatiotemporal comparative analysis of scooter-share and bike-share usage patterns in
Washington, DC. Journal of transport geography, 78, 19-28.
Mi, X., Tang, M., Liao, H., Shen, W., & Lev, B. (2019). The state-of-the-art survey on integrations and
applications of the best worst method in decision making: Why, what, what for and what's
next?. Omega, 87, 205-225.
Millard-Ball, A. (2005). Car-sharing: Where and how it succeeds (Vol. 60). Transportation Research Board.

271
Miller, D.W., and M.K. Starr (1963). Executive Decisions and Operations Research. Prentice-Hall, Inc.,
Englewood Cliffs, NJ,.
Miller, H. E., Thomas, S. L., Smith, K. M., & Robinson, P. (2016). Surveillance, responsibility and control: an
analysis of government and industry discourses about “problem” and “responsible” gambling. Addiction
Research & Theory, 24(2), 163-176.
Miranda-Moreno, L. F., & Nosal, T. (2011). Weather or not to cycle: Temporal trends and impact of weather on
cycling in an urban environment. Transportation research record, 2247(1), 42-52.
Mishra, G. S., Clewlow, R. R., Mokhtarian, P. L., & Widaman, K. F. (2015). The effect of carsharing on vehicle
holdings and travel behavior: A propensity score and causal mediation analysis of the San Francisco Bay
Area. Research in Transportation Economics, 52, 46-55.
Mishra, G. S., Mokhtarian, P. L., Clewlow, R. R., & Widaman, K. F. (2019). Addressing the joint occurrence of
self-selection and simultaneity biases in the estimation of program effects based on cross-sectional
observational surveys: case study of travel behavior effects in carsharing. Transportation, 46(1), 95-123.
Mitchell, G., Tsao, H., Randell, T., Marks, J., & Mackay, P. (2019). Impact of electric scooters to a tertiary
emergency department: 8‐week review after implementation of a scooter share scheme. Emergency
medicine Australasia, 31(6), 930-934.
Mitchell, W. J., Borroni-Bird, C. E., & Burns, L. D. (2010). Reinventing the automobile: Personal urban mobility
for the 21st century. MIT press.
Mitra, R., & Hess, P. M. (2021). Who are the potential users of shared e-scooters? An examination of socio-
demographic, attitudinal and environmental factors. Travel Behaviour and Society, 23, 100-107.
Mohammadi, M., & Rezaei, J. (2020). Bayesian best-worst method: A probabilistic group decision making
model. Omega, 96, 102075.
Moreau, H., de Jamblinne de Meux, L., Zeller, V., D’Ans, P., Ruwet, C., & Achten, W. M. (2020). Dockless e-
scooter: A green solution for mobility? comparative case study between dockless e-scooters, displaced
transport, and personal e-scooters. Sustainability, 12(5), 1803.
Morency, C., Habib, K. M. N., Grasset, V., & Islam, M. T. (2012). Understanding members' carsharing (activity)
persistency by using econometric model. Journal of advanced Transportation, 46(1), 26-38.
Morency, C., Trépanier, M., Agard, B., Martin, B., & Quashie, J. (2007, September). Car sharing system: what
transaction datasets reveal on users' behaviors. In 2007 IEEE Intelligent Transportation Systems
Conference (pp. 284-289). IEEE.
Morency, C., Verreault, H., & Demers, M. (2015). Identification of the minimum size of the shared-car fleet
required to satisfy car-driving trips in Montreal. Transportation, 42(3), 435-447.
Morsche, W. te, La Paix Puello, L., Geurs, K.T. (2019). Potential uptake of adaptive transport services: An
exploration of service attributes and attitudes. Transport Policy, 84, 1–11.
Mota, J., Gomes, H., Almeida, M., Ribeiro, J. C., Carvalho, J., & Santos, M. P. (2007). Active versus passive
transportation to school–differences in screen time, socio-economic position and perceived
environmental characteristics in adolescent girls. Annals of human biology, 34(3), 273-282.
Motoaki, Y., & Daziano, R. A. (2015). A hybrid-choice latent-class model for the analysis of the effects of weather
on cycling demand. Transportation Research Part A: Policy and Practice, 75, 217-230.

272
Mou, Q., Xu, Z., & Liao, H. (2016). An intuitionistic fuzzy multiplicative best-worst method for multi-criteria
group decision making. Information Sciences , 374 , 224-239.
Mounce, R., & Nelson, J. D. (2019). On the potential for one-way electric vehicle car-sharing in future mobility
systems. Transportation Research Part A: Policy and Practice, 120, 17-30.
Mulliner, E., Malys, N., & Maliene, V. (2016). Comparative analysis of MCDM methods for the assessment of
sustainable housing affordability. Omega, 59, 146-156.
Murphy, C. (2016). Shared mobility and the transformation of public transit (No. TCRP J-11/TASK 21).
Murphy, E., & Usher, J. (2015). The role of bicycle-sharing in the city: Analysis of the Irish
experience. International Journal of Sustainable Transportation, 9(2), 116-125.
Namazu, M., & Dowlatabadi, H. (2018). Vehicle ownership reduction: A comparison of one-way and two-way
carsharing systems. Transport Policy, 64, 38-50.
Namazu, M., MacKenzie, D., Zerriffi, H., & Dowlatabadi, H. (2018). Is carsharing for everyone? Understanding
the diffusion of carsharing services. Transport Policy, 63, 189-199.
Nankervis, M. (1999). The effect of weather and climate on bicycle commuting. Transportation Research Part A:
Policy and Practice, 33(6), 417-431.
Nelson, A. C., & Allen, D. (1997). If you build them, commuters will use them: association between bicycle
facilities and bicycle commuting. Transportation research record, 1578(1), 79-83.
Nikitas, A. (2018). Understanding bike-sharing acceptability and expected usage patterns in the context of a small
city novel to the concept: A story of ‘Greek Drama’. Transportation research part F: traffic psychology
and behaviour, 56, 306-321.
Nobis, C. (2006). Carsharing as key contribution to multimodal and sustainable mobility behavior: Carsharing in
Germany. Transportation Research Record, 1986(1), 89-97.
Noble, E. E., & Sanchez, P. P. (1993). A note on the information content of a consistent pairwise comparison
judgment matrix of an AHP decision maker. Theory and Decision, 34(2), 99-108.
Noland, R. B. (2019). Trip patterns and revenue of shared e-scooters in Louisville, Kentucky. Transport Findings.
Noland, R. B., Smart, M. J., & Guo, Z. (2016). Bikeshare trip generation in New York city. Transportation
Research Part A: Policy and Practice, 94, 164-181.
Noland, R. B., Smart, M. J., & Guo, Z. (2019). Bikesharing trip patterns in New York City: Associations with
land use, subways, and bicycle lanes. International Journal of Sustainable Transportation, 13(9), 664-
674.
O’brien, O., Cheshire, J., & Batty, M. (2014). Mining bicycle sharing data for generating insights into sustainable
transport systems. Journal of Transport Geography, 34, 262-273.
Ogilvie, F., & Goodman, A. (2012). Inequalities in usage of a public bicycle sharing scheme: socio-demographic
predictors of uptake and usage of the London (UK) cycle hire scheme. Preventive medicine, 55(1), 40-
45.
Ortúzar, J. de D., Iacobelli, A., & Valeze, C. (2000). Estimating demand for a cycle- way network. Transportation
Research, 34A,353e373.

273
Parkin, J., Wardman, M., & Page, M. (2008). Estimation of the determinants of bicycle mode share for the journey
to work using census data. Transportation, 35(1), 93-109.
Paundra, J., Rook, L., van Dalen, J., & Ketter, W. (2017). Preferences for car sharing services: Effects of
instrumental attributes and psychological ownership. Journal of environmental psychology, 53, 121-130.
Perboli, G., Caroleo, B., Musso, S. (2017). Car-Sharing: Current and Potential Members Behavior Analysis after
the Introduction of the Service. Proceedings - International Computer Software and Applications
Conference, 2, 771–776.
Petersen, A. B. (2019). Scoot over smart devices: The invisible costs of rental scooters. Surveillance &
Society, 17(1/2), 191-197.
Plummer, M. (2004). JAGS: Just another Gibbs sampler.
Popov, A. I., & Ravi, Y. (2020). Conceptualization of service loyalty in access-based services in micromobility:
A case of e-scooter sharing services.
Praditya, D., & Janssen, M. (2017, November). Assessment of factors influencing information sharing
arrangements using the best-worst method. In Conference on e-Business, e-Services and e-Society (pp.
94-106). Springer, Cham.
Prieto, M., Baltas, G., & Stan, V. (2017). Car sharing adoption intention in urban areas: what are the key
sociodemographic drivers?. Transportation Research Part A: Policy and Practice, 101, 218-227.
Pucher, J., & Buehler, R. (2008). Making cycling irresistible: lessons from the Netherlands, Denmark and
Germany. Transport reviews, 28(4), 495-528.
Rahimuddin, M. (2020). Innovation Adoption of New E-Scooters Service in Finland On Consumer Perspective.
Raux, C., Zoubir, A., & Geyik, M. (2017). Who are bike sharing schemes members and do they travel differently?
The case of Lyon’s “Velo’v” scheme. Transportation research part A: policy and practice, 106, 350-363.
Raviv, T., & Kolka, O. (2013). Optimal inventory management of a bike-sharing station. Iie Transactions, 45(10),
1077-1093.
Reiss, S., & Bogenberger, K. (2016). Validation of a relocation strategy for Munich's bike sharing
system. Transportation Research Procedia, 19, 341-349.
Ren, J. (2018). Multi-criteria decision making for the prioritization of energy systems under uncertainties after
life cycle sustainability assessment. Sustainable Production and Consumption, 16, 45-57.
Rezaei, J. (2015). Best-worst multi-criteria decision-making method. Omega, 53, 49-57.
Rezaei, J. (2016). Best-worst multi-criteria decision-making method: Some properties and a linear
model. Omega, 64, 126-130.
Rezaei, J., Arab, A., & Mehregan, M. (2021). Equalizing bias in eliciting attribute weights in multiattribute
decision‐making: experimental research. Journal of Behavioral Decision Making.
Rezaei, J., Hemmes, A., & Tavasszy, L. (2017). Multi-criteria decision-making for complex bundling
configurations in surface transportation of air freight. Journal of Air Transport Management, 61, 95-105.

274
Rezaei, J., Nispeling, T., Sarkis, J., & Tavasszy, L. (2016). A supplier selection life cycle approach integrating
traditional and environmental criteria using the best worst method. Journal of Cleaner Production, 135,
577-588.
Rezaei, J., van Wulfften Palthe, L., Tavasszy, L., Wiegmans, B., & van der Laan, F. (2019). Port performance
measurement in the context of port choice: an MCDA approach. Management Decision.
Rezaei, J., Wang, J., & Tavasszy, L. (2015). Linking supplier development to supplier segmentation using Best
Worst Method. Expert Systems with Applications, 42(23), 9152-9164.
Ricci, M. (2015). Bike sharing: A review of evidence on impacts and processes of implementation and
operation. Research in Transportation Business & Management, 15, 28-38.
Rietveld, P., & Daniel, V. (2004). Determinants of bicycle use: do municipal policies matter?. Transportation
Research Part A: Policy and Practice, 38(7), 531-550.
Riggs, W., & Kawashima, M. (2020). Exploring Best Practice for Municipal E-Scooter Policy in the United
States. Available at SSRN 3512725.
Rixey, R. A. (2013). Station-level forecasting of bikesharing ridership: Station Network Effects in Three US
Systems. Transportation research record, 2387(1), 46-55.
Rodrıguez, D. A., & Joo, J. (2004). The relationship between non-motorized mode choice and the local physical
environment. Transportation Research Part D: Transport and Environment, 9(2), 151-173.
Rotaris, L., & Danielis, R. (2018). The role for carsharing in medium to small-sized towns and in less-densely
populated rural areas. Transportation Research Part A: Policy and Practice, 115, 49-62.
Rotaris, L., Danielis, R., & Maltese, I. (2019). Carsharing use by college students: The case of Milan and
Rome. Transportation Research Part A: Policy and Practice, 120, 239-251.
Roy, B. (1990). The outranking approach and the foundations of ELECTRE methods. In Readings in multiple
criteria decision aid (pp. 155-183). Springer, Berlin, Heidelberg.
Rudloff, C., & Lackner, B. (2014). Modeling demand for bikesharing systems: neighboring stations as source for
demand and reason for structural breaks. Transportation Research Record, 2430(1), 1-11.
Saaty, T. L. (1980). The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation . NY, NY,
USA: McGraw-Hill.
Saaty, T. L. (1994). Fundamentals of Decision Making and Priority Theory, 4922 Ellsworth Ave., Pittsburgh.
Sadaghiani, S., Ahmad, K. W., Rezaei, J., & Tavasszy, L. (2015, April). Evaluation of external forces affecting
supply chain sustainability in oil and gas industry using Best Worst Method. In 2015 International
Mediterranean Gas and Oil Conference (MedGO) (pp. 1-4). IEEE.
Saelens, B. E., Sallis, J. F., & Frank, L. D. (2003). Environmental correlates of walking and cycling: findings from
the transportation, urban design, and planning literatures. Annals of behavioral medicine, 25(2), 80-91.
Safarzadeh, S., Khansefid, S., & Rasti-Barzoki, M. (2018). A group multi-criteria decision-making based on best-
worst method. Computers & Industrial Engineering, 126, 111-121.
Salimi, N. (2017). Quality assessment of scientific outputs using the BWM. Scientometrics, 112(1), 195-213.

275
Salimi, N., & Rezaei, J. (2016). Measuring efficiency of university-industry Ph. D. projects using best worst
method. Scientometrics, 109(3), 1911-1938.
Sallis, J. F., Frank, L. D., Saelens, B. E., & Kraft, M. K. (2004). Active transportation and physical activity:
opportunities for collaboration on transportation and public health research. Transportation Research Part
A: Policy and Practice, 38(4), 249-268.
Salo, A. A., & Hämäläinen, R. P. (1997). On the measurement of preferences in the analytic hierarchy
process. Journal of Multi‐Criteria Decision Analysis, 6(6), 309-319.
Sanders, R. L., Branion-Calles, M., & Nelson, T. A. (2020). To scoot or not to scoot: Findings from a recent
survey about the benefits and barriers of using E-scooters for riders and non-riders. Transportation
Research Part A: Policy and Practice, 139, 217-227.
Saneinejad, S., Roorda, M. J., & Kennedy, C. (2012). Modelling the impact of weather conditions on active
transportation travel behaviour. Transportation research part D: transport and environment, 17(2), 129-
137.
Schaefers, T. (2013). Exploring carsharing usage motives: A hierarchical means-end chain
analysis. Transportation Research Part A: Policy and Practice, 47, 69-77.
Schellong, D., Sadek, P., Schaetzberger, C., & Barrack, T. (2019). The promise and pitfalls of e-scooter
sharing. Europe, 12, 15.
Schlaff, C. D., Sack, K. D., Elliott, R. J., & Rosner, M. K. (2019). Early experience with electric scooter injuries
requiring neurosurgical evaluation in District of Columbia: a case series. World neurosurgery, 132, 202-
207.
Schmöller, S., Weikl, S., Müller, J., & Bogenberger, K. (2015). Empirical analysis of free-floating carsharing
usage: The Munich and Berlin case. Transportation Research Part C: Emerging Technologies, 56, 34-51.
Scholten, L., Maurer, M., & Lienert, J. (2017). Comparing multi-criteria decision analysis and integrated
assessment to support long-term water supply planning. PLoS One, 12(5), e0176663.
Schoner, J., & Levinson, D. M. (2013). Which Station? Access Trips and Bike Share Route Choice.
Schuster, T. D., Byrne, J., Corbett, J., & Schreuder, Y. (2005). Assessing the potential extent of carsharing: A
new method and its implications. Transportation research record, 1927(1), 174-181.
Sener, I. N., Eluru, N., & Bhat, C. R. (2009). An analysis of bicycle route choice preferences in Texas,
US. Transportation, 36(5), 511-539.
Serrai, W., Abdelli, A., Mokdad, L., & Hammal, Y. (2017). Towards an efficient and a more accurate web service
selection using MCDM methods. Journal of computational science, 22, 253-267.
Severengiz, S., Finke, S., Schelte, N., & Wendt, N. (2020, March). Life cycle assessment on the mobility service
E-scooter sharing. In 2020 IEEE European Technology and Engineering Management Summit (E-
TEMS) (pp. 1-6). IEEE.
Shafizadeh, K., & Niemeier, D. (1997). Bicycle journey-to-work: travel behavior characteristics and spatial
attributes. Transportation Research Record, 1578(1), 84-90.
Shaheen, S. A., & Cohen, A. P. (2007). Growth in worldwide carsharing: An international
comparison. Transportation Research Record, 1992(1), 81-89.

276
Shaheen, S. A., & Cohen, A. P. (2013). Carsharing and personal vehicle services: worldwide market developments
and emerging trends. International journal of sustainable transportation, 7(1), 5-34.
Shaheen, S. A., & Lipman, T. E. (2007). Reducing greenhouse emissions and fuel consumption: Sustainable
approaches for surface transportation. IATSS research, 31(1), 6-20.
Shaheen, S. A., & Martin, E. (2010). Demand for carsharing systems in Beijing, China: an exploratory
study. International journal of sustainable transportation, 4(1), 41-55.
Shaheen, S. A., & Rodier, C. J. (2005). Travel effects of a suburban commuter carsharing service: CarLink case
study. Transportation research record, 1927(1), 182-188.
Shaheen, S. A., Chan, N. D., & Micheaux, H. (2015). One-way carsharing’s evolution and operator perspectives
from the Americas. Transportation, 42(3), 519-536.
Shaheen, S. A., Cohen, A. P., & Roberts, J. D. (2006). Carsharing in North America: Market growth, current
developments, and future potential. Transportation Research Record, 1986(1), 116-124.
Shaheen, S. A., Guzman, S., & Zhang, H. (2010). Bikesharing in Europe, the Americas, and Asia: past, present,
and future. Transportation Research Record, 2143(1), 159-167.
Shaheen, S. A., Sperling, D., & Wagner, C. (1999). A Short History of Carsharing in the 90's.
Shaheen, S. A., Zhang, H., Martin, E., & Guzman, S. (2011). China's Hangzhou public bicycle: understanding
early adoption and behavioral response to bikesharing. Transportation Research Record, 2247(1), 33-41.
Shaheen, S., & Chan, N. (2016). Mobility and the sharing economy: Potential to facilitate the first-and last-mile
public transit connections. Built Environment, 42(4), 573-588.
Shaheen, S., & Cohen, A. (2019). Shared Micromoblity Policy Toolkit: Docked and Dockless Bike and Scooter
Sharing.
Shaheen, S., & Cohen, A. (2020). Innovative Mobility: Carsharing Outlook Carsharing Market Overview,
Analysis, And Trends.
Shaheen, S., & Guzman, S. (2011). Worldwide bikesharing. Access Magazine, 1(39), 22-27.
Shaheen, S., & Wright, J. (2001, August). The Carlink II pilot program: Testing A commuter-based carsharing
model. In ITSC 2001. 2001 IEEE Intelligent Transportation Systems. Proceedings (Cat. No.
01TH8585) (pp. 1067-1072). IEEE.
Shaheen, S., Cohen, A., & Zohdy, I. (2016). Shared mobility: current practices and guiding principles (No.
FHWA-HOP-16-022). United States. Federal Highway Administration.
Shaheen, S., Cohen, A., Chan, N., & Bansal, A. (2020). Sharing strategies: carsharing, shared micromobility
(bikesharing and scooter sharing), transportation network companies, microtransit, and other innovative
mobility modes. In Transportation, Land Use, and Environmental Planning (pp. 237-262). Elsevier.
Shaheen, S., Guzman, S., & Zhang, H. (2012). Bikesharing across the globe. City cycling, 183.
Shaheen, S., Martin, E., & Bansal, A. (2018). Peer-To-Peer (P2P) carsharing: Understanding early markets, social
dynamics, and behavioral impacts.
Shaheen, S., Wright, J., Dick, D., & Novick, L. (2000). CarLink-A smart carsharing system field test report.

277
Sharma, Y. K., Mangla, S. K., & Patil, P. P. (2019). Analyzing challenges to transportation for successful
sustainable food supply chain management implementation in Indian dairy industry. In Information and
communication technology for competitive strategies (pp. 409-418). Springer, Singapore.
Shen, Y., Zhang, X., & Zhao, J. (2018). Understanding the usage of dockless bike sharing in
Singapore. International Journal of Sustainable Transportation, 12(9), 686-700.
Si, H., Shi, J. G., Wu, G., Chen, J., & Zhao, X. (2019). Mapping the bike sharing research published from 2010
to 2018: A scientometric review. Journal of cleaner production, 213, 415-427.
Simons, D., Clarys, P., De Bourdeaudhuij, I., de Geus, B., Vandelanotte, C., & Deforche, B. (2013). Factors
influencing mode of transport in older adolescents: a qualitative study. BMC public health, 13(1), 323.
Sioui, L., Morency, C., & Trépanier, M. (2013). How carsharing affects the travel behavior of households: a case
study of montréal, Canada. International journal of sustainable transportation, 7(1), 52-69.
Sipe, N., & Pojani, D. (2018). Can e-scooters solve the 'last mile' problem? They'll need to avoid the fate of
dockless bikes.
Skov-Petersen, H., Jacobsen, J. B., Vedel, S. E., Alexander, S. N. T., & Rask, S. (2017). Effects of upgrading to
cycle highways-an analysis of demand induction, use patterns and satisfaction before and after. Journal
of transport geography, 64, 203-210.
Smith, C. S., & Schwieterman, J. P. (2018). E-scooter scenarios: evaluating the potential mobility benefits of
shared dockless scooters in Chicago.
Spieser, K., Samaranayake, S., Gruel, W., & Frazzoli, E. (2016, January). Shared-vehicle mobility-on-demand
systems: A fleet operator’s guide to re-balancing empty vehicles. In Transportation Research Board 95th
Annual Meeting (No. 16-5987). Transportation Research Board.
Standing, C., Standing, S., & Biermann, S. (2019). The implications of the sharing economy for
transport. Transport Reviews, 39(2), 226-242.
Stević, Ž., Pamučar, D., Subotić, M., Antuchevičiene, J., & Zavadskas, E. K. (2018). The location selection for
roundabout construction using Rough BWM-Rough WASPAS approach based on a new Rough Hamy
aggregator. Sustainability, 10(8), 2817.
Stillwater, T., Mokhtarian, P. L., & Shaheen, S. (2008). Carsharing and the built environment: A GIS-based study
of one US operator. Institute of Transportation Studies.
Strong, W. B., Malina, R. M., Blimkie, C. J., Daniels, S. R., Dishman, R. K., Gutin, B., ... & Rowland, T. (2005).
Evidence based physical activity for school-age youth. The Journal of pediatrics, 146(6), 732-737.
Sun, F., Chen, P., & Jiao, J. (2018). Promoting public bike-sharing: A lesson from the unsuccessful Pronto
system. Transportation Research Part D: Transport and Environment, 63, 533-547.
Sun, Y. (2018). Sharing and riding: How the dockless bike sharing scheme in China shapes the city. Urban
Science, 2(3), 68.
Tang, Y., Pan, H., & Shen, Q. (2011, January). Bike-sharing systems in Beijing, Shanghai, and Hangzhou and
their impact on travel behavior. In 90th Annual Meeting of the Transportation Research Board,
Washington, DC.

278
Ter Schure, J., Napolitan, F., & Hutchinson, R. (2012). Cumulative impacts of carsharing and unbundled parking
on vehicle ownership and mode choice. Transportation research record, 2319(1), 96-104.
Terrien, C., Maniak, R., Chen, B., & Shaheen, S. (2016). Good practices for advancing urban mobility innovation:
A case study of one-way carsharing. Research in transportation business & management, 20, 20-32.
Tilahun, N. Y., Levinson, D. M., & Krizek, K. J. (2007). Trails, lanes, or traffic: Valuing bicycle facilities with
an adaptive stated preference survey. Transportation Research Part A: Policy and Practice, 41(4), 287-
301.
Todd, J., Krauss, D., Zimmermann, J., & Dunning, A. (2019). Behavior of electric scooter operators in naturalistic
environments (No. 2019-01-1007). SAE Technical Paper.
Torabi, S. A., Giahi, R., & Sahebjamnia, N. (2016). An enhanced risk assessment framework for business
continuity management systems. Safety science, 89, 201-218.
Triantaphyllou, E. (2000). Multi-criteria decision making methods. In Multi-criteria decision making methods: A
comparative study (pp. 5-21). Springer, Boston, MA.
Trivedi, B., Kesterke, M. J., Bhattacharjee, R., Weber, W., Mynar, K., & Reddy, L. V. (2019). Craniofacial injuries
seen with the introduction of bicycle-share electric scooters in an urban setting. Journal of Oral and
Maxillofacial Surgery, 77(11), 2292-2297.
Trivedi, T. K., Liu, C., Antonio, A. L. M., Wheaton, N., Kreger, V., Yap, A., ... & Elmore, J. G. (2019). Injuries
associated with standing electric scooter use. JAMA network open, 2(1), e187381-e187381.
Tubis, A., Rydlewski, M., & Skupień, E. (2019, September). Non-Technical Aspects of Safety in Scooter-Sharing
System in Wroclaw. In Scientific And Technical Conference Transport Systems Theory And
Practice (pp. 163-173). Springer, Cham.
Tummala, V. R., & Wan, Y. W. (1994). On the mean random inconsistency index of analytic hierarchy process
(AHP). Computers & industrial engineering, 27(1-4), 401-404.
Tuncer, S., & Brown, B. (2020, April). E-scooters on the Ground: Lessons for Redesigning Urban Micro-Mobility.
In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (pp. 1-14).
Tuncer, S., Laurier, E., Brown, B., & Licoppe, C. (2020). Notes on the practices and appearances of e-scooter
users in public space. Journal of transport geography, 85, 102702.
Turoń, K., & Czech, P. (2019, September). The Concept of Rules and Recommendations for Riding Shared and
Private E-Scooters in the Road Network in the Light of Global Problems. In Scientific And Technical
Conference Transport Systems Theory And Practice (pp. 275-284). Springer, Cham.
Turoń, K., Kubik, A., & Chen, F. (2021). Electric Shared Mobility Services during the Pandemic: Modeling
Aspects of Transportation. Energies, 14(9), 2622.
Turoń, K., Kubik, A., Chen, F., Wang, H., & Łazarz, B. (2020). A holistic approach to electric shared mobility
systems development—Modelling and optimization aspects. Energies, 13(21), 5810.
Turoń, K., Kubik, A., Chen, F., Wang, H., & Łazarz, B. (2020). A holistic approach to electric shared mobility
systems development—Modelling and optimization aspects. Energies, 13(21), 5810.

279
Tuzkaya, G., Gülsün, B., Kahraman, C., & Özgen, D. (2010). An integrated fuzzy multi-criteria decision making
methodology for material handling equipment selection problem and an application. Expert systems with
applications, 37(4), 2853-2863.
Uteng, T. P., Julsrud, T. E., & George, C. (2019). The role of life events and context in type of car share uptake:
Comparing users of peer-to-peer and cooperative programs in Oslo, Norway. Transportation Research
Part D: Transport and Environment, 71, 186-206.
Vahdani, B., Jabbari, A. H. K., Roshanaei, V., & Zandieh, M. (2010). Extension of the ELECTRE method for
decision-making problems with interval weights and data. The International Journal of Advanced
Manufacturing Technology, 50(5-8), 793-800.
Vahidi, F., Torabi, S. A., & Ramezankhani, M. J. (2018). Sustainable supplier selection and order allocation under
operational and disruption risks. Journal of Cleaner Production, 174, 1351-1365.
Vinayak, P., Dias, F. F., Astroza, S., Bhat, C. R., Pendyala, R. M., & Garikapati, V. M. (2018). Accounting for
multi-dimensional dependencies among decision-makers within a generalized model framework: An
application to understanding shared mobility ser
Vogel, M., Hamon, R., Lozenguez, G., Merchez, L., Abry, P., Barnier, J., ... & Robardet, C. (2014). From bicycle
sharing system movements to users: a typology of Vélo’v cyclists in Lyon based on large-scale
behavioural dataset. Journal of Transport Geography, 41, 280-291.
Vogel, P., & Mattfeld, D. C. (2011, September). Strategic and operational planning of bike-sharing systems by
data mining–a case study. In International conference on computational logistics (pp. 127-141). Springer,
Berlin, Heidelberg.
Vogel, P., Greiser, T., & Mattfeld, D. C. (2011). Understanding bike-sharing systems using data mining:
Exploring activity patterns. Procedia-Social and Behavioral Sciences, 20, 514-523.
Vosooghi, R., Puchinger, J., Jankovic, M., & Sirin, G. (2017, October). A critical analysis of travel demand
estimation for new one-way carsharing systems. In 2017 IEEE 20th International Conference on
Intelligent Transportation Systems (ITSC) (pp. 199-205). IEEE.
Vulević, T., & Dragović, N. (2017). Multi-criteria decision analysis for sub-watersheds ranking via the
PROMETHEE method. International Soil and Water Conservation Research, 5(1), 50-55.
Wagner, S., Brandt, T., & Neumann, D. (2015, January). Data analytics in free-floating carsharing: Evidence from
the city of berlin. In 2015 48th Hawaii International Conference on System Sciences (pp. 897-907).
IEEE.
Wagner, S., Brandt, T., & Neumann, D. (2016). In free float: Developing Business Analytics support for car-
sharing providers. Omega, 59, 4-14.
Waldman, J. A. (1977). Cycling in towns. quantitative investigation (No. LTR Working Paper 3 Monograph).
Wang, J., & Lindsey, G. (2019). Do new bike share stations increase member use: A quasi-experimental
study. Transportation research part A: policy and practice, 121, 1-11.
Wang, K., & Akar, G. (2019). Gender gap generators for bike share ridership: Evidence from Citi Bike system in
New York City. Journal of transport geography, 76, 1-9.

280
Wang, K., Akar, G., & Chen, Y. J. (2018). Bike sharing differences among millennials, Gen Xers, and baby
boomers: Lessons learnt from New York City’s bike share. Transportation research part A: policy and
practice, 116, 1-14.
Wang, M., Martin, E. W., & Shaheen, S. A. (2012). Carsharing in Shanghai, China: analysis of behavioral
response to local survey and potential competition. Transportation research record, 2319(1), 86-95.
Wang, X., Lindsey, G., Schoner, J. E., & Harrison, A. (2016). Modeling bike share station activity: Effects of
nearby businesses and jobs on trips to and from stations. Journal of Urban Planning and
Development, 142(1), 04015001.
Wang, Y. (2015). Evaluation Of Free Bicycle Project In Wuhan, China.
Wang, Y. W. (2008). Simulation of service capacity an electric scooter refueling system. Transportation Research
Part D: Transport and Environment, 13(2), 126-132.
Wang, Y., Yan, X., Zhou, Y., Xue, Q., & Sun, L. (2017). Individuals’ acceptance to free-floating electric
carsharing mode: A web-based survey in China. International journal of environmental research and
public health, 14(5), 476.
Wardman, M., Tight, M., & Page, M. (2007). Factors influencing the propensity to cycle to work. Transportation
Research Part A: Policy and Practice, 41(4), 339-350.
Watson, H. N., Garman, C. M., Wishart, J., & Zimmermann, J. (2020). Patient Demographics and Injury
Characteristics of ER Visits Related to Powered-Scooters (No. 2020-01-0933). SAE Technical Paper.
Weikl, S., & Bogenberger, K. (2013). Relocation strategies and algorithms for free-floating car sharing
systems. IEEE Intelligent Transportation Systems Magazine, 5(4), 100-111.
Welch, T. F., Gehrke, S. R., & Widita, A. (2020). Shared-use mobility competition: a trip-level analysis of taxi,
bikeshare, and transit mode choice in Washington, DC. Transportmetrica A: transport science, 16(1), 43-
55.
Whalen, K. E., Páez, A., & Carrasco, J. A. (2013). Mode choice of university students commuting to school and
the role of active travel. Journal of Transport Geography, 31, 132-142.
Wielinski, G., Trépanier, M., & Morency, C. (2015). What about free-floating carsharing? A look at the Montreal,
Canada, case. Transportation Research Record, 2563(1), 28-36.
Winter, K., Cats, O., Martens, K., & van Arem, B. (2017). A stated-choice experiment on mode choice in an era
of free-floating carsharing and shared autonomous vehicles (No. 17-01321).
Winters, M., Brauer, M., Setton, E. M., & Teschke, K. (2010). Built environment influences on healthy
transportation choices: bicycling versus driving. Journal of urban health, 87(6), 969-993.
Winters, M., Davidson, G., Kao, D., & Teschke, K. (2011). Motivators and deterrents of bicycling: comparing
influences on decisions to ride. Transportation, 38(1), 153-168.
Wüster, J., Voß, J., Koerdt, S., Beck-Broichsitter, B., Kreutzer, K., Märdian, S., ... & Doll, C. (2020). Impact of
the Rising Number of Rentable E-scooter Accidents on Emergency Care in Berlin 6 Months after the
Introduction: A Maxillofacial Perspective. Craniomaxillofacial Trauma & Reconstruction,
1943387520940180.

281
Xing, Y., Handy, S. L., & Mokhtarian, P. L. (2010). Factors associated with proportions and miles of bicycling
for transportation and recreation in six small US cities. Transportation research part D: Transport and
Environment, 15(2), 73-81.
Xu, S. J., & Chow, J. Y. (2019). A longitudinal study of bike infrastructure impact on bikesharing system
performance in New York City. International Journal of Sustainable Transportation, 1-16.
Yannis, G., Kopsacheili, A., Dragomanovits, A., & Petraki, V. (2020). State-of-the-art review on multi-criteria
decision-making in the transport sector. Journal of traffic and transportation engineering (English
edition), 7(4), 413-431.
Yoon, T., Cherry, C. R., & Jones, L. R. (2017). One-way and round-trip carsharing: A stated preference
experiment in Beijing. Transportation research part D: transport and environment, 53, 102-114.
Younes, H., Zou, Z., Wu, J., & Baiocchi, G. (2020). Comparing the temporal determinants of dockless scooter-
share and station-based bike-share in Washington, DC. Transportation Research Part A: Policy and
Practice, 134, 308-320.
Zahran, S., Brody, S. D., Maghelal, P., Prelog, A., & Lacy, M. (2008). Cycling and walking: Explaining the spatial
distribution of healthy modes of transportation in the United States. Transportation research part D:
transport and environment, 13(7), 462-470.
Zanotto, M. (2014). Facilitators and barriers to public bike share adoption and success in a city with compulsory
helmet legislation: A mixed-methods approach (Doctoral dissertation, Health Sciences: Faculty of Health
Sciences).
Zhang, H., Yin, C., Qi, X., Zhang, R., & Kang, X. (2017). Cognitive best worst method for multiattribute decision-
making. Mathematical Problems in Engineering, 2017.
Zhang, L., Zhang, J., Duan, Z. Y., & Bryde, D. (2015). Sustainable bike-sharing systems: characteristics and
commonalities across cases in urban China. Journal of Cleaner Production, 97, 124-133.
Zhang, Y. (2017). Bike-sharing Usage: Mining on the Trip Data of Bike-sharing Users. University of Twente
Faculty of Geo-Information and Earth Observation (ITC).
Zhang, Y., & Mi, Z. (2018). Environmental benefits of bike sharing: A big data-based analysis. Applied
Energy, 220, 296-301.
Zhao, D., Ong, G. P., Wang, W., & Hu, X. J. (2019). Effect of built environment on shared bicycle reallocation:
A case study on Nanjing, China. Transportation research part A: policy and practice, 128, 73-88.
Zheng, J., Scott, M., Rodriguez, M., Sierzchula, W., Platz, D., Guo, J. Y., & Adams, T. M. (2009). Carsharing in
a university community: Assessing potential demand and distinct market characteristics. Transportation
research record, 2110(1), 18-26.
Zhou, B., & Kockelman, K. M. (2011). Opportunities for and impacts of carsharing: A survey of the Austin, Texas
market. International Journal of Sustainable Transportation, 5(3), 135-152.
Zhou, F., Zheng, Z., Whitehead, J., Washington, S., Perrons, R. K., & Page, L. (2020). Preference heterogeneity
in mode choice for car-sharing and shared automated vehicles. Transportation Research Part A: Policy
and Practice, 132, 633-650.
Zhou, X. (2015). Understanding spatiotemporal patterns of biking behavior by analyzing massive bike sharing
data in Chicago. PloS one, 10(10), e0137922.

282
Zhu, R., Zhang, X., Kondor, D., Santi, P., & Ratti, C. (2020). Understanding spatio-temporal heterogeneity of
bike-sharing and scooter-sharing mobility. Computers, Environment and Urban Systems, 81, 101483.
Zoepf, S. M., & Keith, D. R. (2016). User decision-making and technology choices in the US carsharing
market. Transport Policy, 51, 150-157.
Zou, Z., Younes, H., Erdoğan, S., & Wu, J. (2020). Exploratory Analysis of Real-Time E-Scooter Trip Data in
Washington, DC. Transportation Research Record, 0361198120919760.

283
Appendix 1
Appendix 1: Details on the methodology
of the review of the socio-demographic
factors for car-sharing and previous
reviews in the same area
16
This section details the methodological steps of reviewing the socio-demographic factors that
can affect demand for different car-sharing forms presented in chapter 2. Similar steps have
also been taken to review those factors influencing bike-sharing and scooter-sharing usage.
Previous work has already reviewed these factors. Jorge and Correia (2013) examined
research that developed models to describe car-sharing demand and focused on solving the
problem of vehicle imbalance. Ferrero et al. (2018) categorized the research, identified
mainstream, and studied trends and perspectives. Illgen and H¨ock (2019) reviewed the papers
that provided solutions to car-sharing relocation problems in the networks. Besides, Liao and
Correia (2020) reviewed the publications that focused on demand estimation, use patterns, and
potential impacts of Electric Car-Sharing (E-Car-Sharing). Unlike many previous studies that
often did not explicitly consider different car-sharing variants, it is explicitly acknowledged
that the operational scheme can profoundly impact the targeted travel demand segment.
Therefore, an effort is made in the following analysis to distinguish the impacts of passengers’
socio-demographic characteristics on different shared car schemes.
The following are the steps taken in this study to complete the review, mainly based on
the method presented in similar studies (Akter et al., 2021; Eren and Uz, 2020; Nguyen et al.,
2021; Rand and Fleming, 2019; Sadri et al., 2021). For a review, four databases, including
16
Most of the contents of the present appendix have been published in Amirnazmiafshar, E., & Diana, M. (2022).
A review of the socio-demographic characteristics affecting the demand for different car-sharing operational
schemes. Transportation Research Interdisciplinary Perspectives, 14, 100616.

284
Google Scholar, TRID (https://trid.trb.org/), Scopus, and Web of Science, were used to
evaluate recent papers on the car-sharing system according to a keywords-based process.
During this process, no lower bounds on the publication date of reviewed papers are
considered. The upper bound is December 22, 2020.
Several searches are performed in the mentioned databases by combining the keywords
related to shared car systems like socio-demographic characteristics, demand for car-sharing,
and car-sharing programs. These keywords were combined to form the set of strings used in
the search, as listed by rows in Table A1.
Table A1: Number of selected articles by each keyword in each database.
String of Keywords
Google
Scholar
TRID
Scopus
Web of
Science
Total (With
Duplicates)
Impacts of carsharing
44
9
4
31
88
Carsharing demand
43
3
13
30
89
Carsharing use
54
7
8
28
97
Gender effect on carsharing use
52
1
-
1
54
Sociodemographic factors' effects on
carsharing
33
-
-
-
33
Users’ behavior of a carsharing
31
10
-
26
67
Carsharing attraction
28
-
-
1
29
Carsharing adoption
19
6
6
11
42
Total (with duplicates)
304
36
31
128
499
For each keyword, the title of the first 100 articles (if any) of each database was reviewed,
totaling 1979 articles. As indicated in Table A1, 499 articles were selected based on titles that,
at first glance, seemed relevant to the purpose of the study. After eliminating duplicates (354
articles), 145 articles remained. An additional set of 23 articles was reviewed, including articles
cited in the articles obtained by keywords, articles selected based on the author’s knowledge,
and articles used to explain the methodology of this article. These additional articles were not
among the 499 articles because they did not contain the abovementioned keywords. Therefore,
this initial pool of published papers consisted of 145 + 23 = 168 articles.
This pool was then scanned to select those focusing on different car-sharing systems’
features and important factors influencing the service demand. Hence, 64 articles were not
considered in this study since they mainly covered topics such as the benefits of the shared car,
history and car-sharing trends, car-sharing classification, interaction with other modes of
transport, and re-balancing issues. Therefore, 104 articles were left. Additionally, 13 articles
not significantly dealing with the socio-demographic effects on demand for car-sharing
services were discarded. These features included the trip purpose (2 articles), trip distance (2
articles), travel time (1 article), travel distance (1 article), Provision of Electric Vehicles (1
article), land use (2 articles), accessibility, and fleet size (3 articles), travel cost (1 article) that
were omitted. Six of these 13 discarded articles focused on more than one non-socio-
demographic feature.
In total, 91 articles have then been considered in this review paper coming from 25
different journals, two different conference proceedings, and four from research or educational
reports. Among these articles, 59 directly mentioned the socio-demographic characteristics

285
influencing car-sharing demand. The other 32 articles were not discarded and were used to
cover other sections of the article, such as the introduction and method.
Given the uneven attention that previous research has paid to different characteristics,
the conclusions or claims of the present review are based on only a few studies in some cases,
while several studies have been reviewed for other claims. Figure A1 illustrates the number of
studies examined for each of the eight socio-demographic characteristics according to the type
of car-sharing services. Therefore, it helps to analyze and understand the degree of support for
some of the results. It should be noted that the “station-based (type is not specified)” in Figure
A1 refers to articles that did not explicitly state whether the authors worked on round-trip
station-based car-sharing or one-way station-based car-sharing. It is only mentioned that they
have worked on station-based car-sharing. “round-trip” in Figure A1 refers to articles working
on home zone-based round-trip car-sharing or station-based round-trip car-sharing. Obviously,
also differences in findings among studies on the same issue should be considered to assess if
such findings are well established.
Figure A1: The number of studies reviewed for each socio-demographic characteristic
according to the type of car-sharing service.
3
6
32 2
01
3
5
7 7
5
2
0
4
7
2323
1 1 1
4
121 1 0 0 0
2
3 3
54
0
2 2
5
1 1 210 0 0 0
0
1
2
3
4
5
6
7
8
Free-floating More than one type of car-sharing
Station-based (type is not specified) One-way station-based
Round trip P2P

286
Appendix 2
Appendix 2: Survey questionnaires
In this study, nine surveys are used to understand the perspectives of four stakeholders of three
shared mobility services, including car-sharing, bike-sharing, and scooter-sharing
(individually), as well as shared mobility services (as a whole, not for a specific shared mobility
service). These nine surveys, numbered from 1 to 9, are listed as follows.
• Survey 1: users and non-users of car-sharing services
• Survey 2: users and non-users of bike-sharing services
• Survey 3: users and non-users of scooter-sharing services
• Survey 4: government members and operators of car-sharing services
• Survey 5: government members and operators of bike-sharing services
• Survey 6: government members and operators of scooter-sharing services
• Survey 7: users and non-users of shared mobility services (as a whole, not for a
specific shared mobility service)
• Survey 8: government members who respond to the shared mobility services (as a
whole, not for a specific shared mobility service) surveys
• Survey 9: operators of shared mobility services (as a whole, not for a specific shared
mobility service).
In these surveys, additional explanations are written in italics for the company
conducting the survey. Also, the question filters (for example, if question "1" is yes, then
answer this question) are written in italics. It is important to note that surveys (surveys 1 to 3)
of users and non-users were the same. Also, government members (surveys 4 to 6) and
operators (surveys 4 to 6) answered identical surveys for each shared mobility service. For the
four stakeholders, the BWM-related questions (question set A in surveys 1 to 6) were the same
(to understand the difference in their views on the same factors). Still, the rest of the questions
users and non-users asked differed from those of government members and operators. In the
surveys of car-sharing, bike-sharing, and scooter-sharing (surveys 1 to 6), the three different
variants in the wording of some questions are reported as "{car, bike, scooter}-sharing". Also,

287
there is a slight difference in the options of some of the questions for different shared mobility
services (car-sharing, bike-sharing, and scooter-sharing), which are marked, for example, as
follows: “•{Car-sharing questionnaire only: Driver}”. Therefore, in sub-section A2.1 and A2.2,
two surveys are presented separately for users and non-users stakeholders (surveys 1 to 3) and
the government members and operators stakeholders (surveys 4 to 6), respectively.
Furthermore, the type of surveys conducted among government members (survey 8) and
operators (survey 9) for shared transportation services (as a whole, not for a specific shared
mobility service) was different because the purpose was to understand the importance of factors
related to their decision, which were different for these two groups (question set A). Hence,
three surveys for users/non-users (survey 7), government members (survey 8), and operators
(survey 9) of shared mobility services (as a whole, not for a specific shared mobility service)
are presented separately in the following three subsections A2.3, A2.4, and A2.5, respectively.
It should be noted that all surveys (surveys 1 to 9) were administered in Italian, even if an
English translation is reported here.
A2.1 Questionnaires for users and non-users of each shared mobility service
(surveys 1 to 3)
This type of survey (1 to 3) is designed for users and non-users of car sharing, bike-sharing,
and scooter-sharing services, and it includes two parts. In the first part, there are questions
related to BWM analysis. In the second part, there are questions relevant to the routines, daily
travel views, and socio-economic situation of respondents.
Dear Ms./Mr.
We are conducting a study at Politecnico di Torino. We aim to understand better your views on the
importance of different characteristics in {car, bike, scooter}-sharing, to know your mobility routines and
your daily travels. We assure you that all information you provide will be treated with the utmost
confidentiality and will be completely anonymous. Your participation is a valuable contribution to this
study, and we thank you for your cooperation.
Please read the {car, bike, scooter}-sharing definition first.
{car, bike, scooter}-sharing definition: People can use {car, bike, scooter}-sharing in many cities and
communities. As a user, you have access to bookable {car, bike, scooter}-sharing vehicles. The vehicles are
available 24 hours a day, 7 days a week, and available through self-service. It is important to note that a
trip through {car, bike, scooter}-sharing is not shared with other users, but it is only the vehicles that are
shared with others who use them at other times.
Please answer the following questions (question set A).
B1. There are several trip-related characteristics that could be considered in selecting {car, bike, scooter}-
sharing to make a trip. These characteristics are listed below.
• Travel time: the time it takes with a given means to travel from origin to destination.
• Travel distance: the distance between origin and destination.
• Departure time: the trip's start time, such as in the morning or evening, on weekends, or on weekdays,
during peak or off-peak hours.
• Trip purpose: the purpose of the trip, such as traveling to work, school, shopping, or meeting a friend.
In your opinion, what is the MOST IMPORTANT, and what is the LEAST IMPORTANT trip-related
characteristic among the above four that could drive your choice?

288
Trip-related
characteristics
Select the most important characteristic in the cell
below
Select the least important characteristic in the cell
below
Travel time
Travel distance
Departure time
Trip purpose
“4*2=8 radio buttons in the above table are needed to make the selections”.
B2. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other three characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important
factor and the other three should be mentioned in the first column according to the template below. 9*3 = 27
radio buttons should appear in the table. Characteristic 1 is always "the least important characteristic" the
respondent selected in the previous step.
To which extent MOST IMPORTANT is more important
than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
Characteristic 2
Characteristic 3
B3. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other two characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other two characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*2 = 18 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely more
important
9
Characteristic 2 is more important than LEAST
IMPORTANT
Characteristic 3 is more important than LEAST
IMPORTANT
B4. Now, let us examine the relative importance of some {car, bike, scooter}-sharing characteristics. These
characteristics are listed below.
• Cost: expenses for {car, bike, scooter}-sharing usage such as service subscription fees or usage fees.
• Comfort: vehicle characteristics that make you feel comfortable during the trip.
• Safety: The level of safety of the people during the trip, such as the rate of accidents, harassment, assault,
and theft.
• Service quality: Quality of the {car, bike, scooter}-sharing system and given services.
• Environment-friendly system: a system that is reducing environmental impacts.
• User-friendliness: easy for beginners to learn, easy to use, and provide travel information in the app.
In your opinion, what is the MOST IMPORTANT, and what is the LEAST IMPORTANT {car, bike, scooter}-
sharing characteristic among the above six that could drive your choice?
{Car, bike, scooter}-
sharing characteristics
Select the most important characteristic in the
cell below
Select the least important characteristic in the
cell below
Travel cost
Travel comfort
Safety
Service quality
Environment-friendly
system
User-friendly
“6*2=12 radio buttons in the above table are needed to make the selections
B5. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other five characteristics?

289
The respondent should see the following table, where the characteristic which is selected as the most important
and the other five should be mentioned in the first column according to the template below. 9*5 = 45 radio buttons
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST IMPORTANT is more important
than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely more
important
9
Least important)
Characteristic 2
Characteristic 3
Characteristic 4
Characteristic 5
B6. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other four characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other four characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*4 = 36 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely more
important
9
Characteristic 2 is more important than LEAST
IMPORTANT
Characteristic 3 is more important than LEAST
IMPORTANT
Characteristic 4 is more important than LEAST
IMPORTANT
Characteristic 5 is more important than LEAST
IMPORTANT
B7. Finally, let us consider the following two characteristics related to where shared cars are actually available:
• Service availability: Availability of {car, bike, scooter}-sharing services around shopping malls,
colleges, transportation centers, city centers, and densely populated areas.
• Vehicle availability and accessibility: Availability of the vehicle where I need it, easiness to reach and
access the vehicle, proximity to the location of the parked vehicle from my starting point.
In your opinion, what is the MOST IMPORTANT factor between these two?
• Service availability
• Vehicle availability and accessibility
B8. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than LEAST IMPORTANT?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other one should be mentioned in the first column according to the template below. 9*1 = 9 radio buttons
should appear in the table.
To which extent MOST IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely more
important
9
LEAST IMPORTANT
B9. Now, let us jointly consider trip-related characteristics, {car, bike, scooter}-sharing characteristics, and
availability & accessibility that you separately assessed in the previous questions. In your opinion, which of these
three sets of characteristics is overall the MOST IMPORTANT, and which is the LEAST IMPORTANT when
considering selecting {car, bike, scooter}-sharing to make a trip?
Characteristics
Select the most important
characteristic in the cell below
Select the least important
characteristic in the cell below

290
Trip-related characteristics (travel time, travel
distance, departure time, trip purpose)
{car, bike, scooter}-sharing characteristics
(Cost, comfort, safety, service quality,
environment-friendly system, user-friendliness)
Availability and accessibility
(Service availability, vehicle availability and
accessibility)
3*2=6 radio buttons in the above table are needed to make the selections.
B10. In the above question, you have chosen MOST_IMPORTANT as the most important set of characteristics.
Could you please rate to which extent you consider MOST_IMPORTANT more important than the other two sets?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other two should be mentioned in the first column according to the template below. 9*2 = 18 radio buttons
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST IMPORTANT is more important
than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely more
important
9
Least important)
Characteristic 2
B11. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other characteristic more important than LEAST_IMPORTANT?
The respondent should see the following table, where the other characteristic, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the below template. 9*1 = 9 radio
buttons should appear in the table.
To which extent…
Equal
Importance
1
2
3
4
5
6
7
8
Extremely more
important
9
Characteristic 2 is more important than LEAST
IMPORTANT
The following questions are about your routines and your daily travel views (question set B).
Q1. Do you have a driving license?
• Yes
• No
Q2. Do you have any experience with {car, bike, scooter}-sharing services?
1. Yes, I am currently using {car, bike, scooter}-sharing services.
2. Yes, I used {car, bike, scooter}-sharing in the past, but I no longer use it.
3. No, I never used {car, bike, scooter}-sharing, but I know what it is.
4. I am not familiar with the concept of {car, bike, scooter}-sharing.
If 1, 2, or 3 in Q2
Q3. To what extent are you familiar with {car, bike, scooter}-sharing? (Membership terms, how to book, price
levels, etc.)
• 1 (Slightly familiar)
• 2
• 3
• 4
• 5 (Very Familiar)
If 1 in Q2

291
Q4. Which {car, bike, scooter}-sharing services do you use? (Show list of {car, bike, scooter}-sharing operators
in Turin) (Respondents can choose more than one option at a time)
• {Car-sharing questionnaire only: Enjoy}
• {Car-sharing questionnaire only: Car2go (Share Now)}
• {Car-sharing questionnaire only: BlueTorino}
• {Bike-sharing questionnaire only: TOBike}
• {Bike-sharing questionnaire only: Mobike}
• {Scooter-sharing questionnaire only: Bird}
• {Scooter-sharing questionnaire only: BIT mobility}
• {Scooter-sharing questionnaire only: Dott}
• {Scooter-sharing questionnaire only: Helbiz An}
• {Scooter-sharing questionnaire only: Circ}
• {Scooter-sharing questionnaire only: Lime}
• {Scooter-sharing questionnaire only: Wind}
• {Scooter-sharing questionnaire only: Link}
• {Scooter-sharing questionnaire only: Vo i}
• Other (please specify)……
If 1, 2, or 3 in Q2
Q5. Are there any {car, bike, scooter}-sharing pick-up locations near your home, or is your home within an
operational area of at least one {car, bike, scooter}-sharing service?
• Yes
• No
• I do not know.
If 1, 2, or 3 in Q2
Q6. Are there any {car, bike, scooter}-sharing pick-up locations near the most frequent destination of your trips
(e.g., workplace, the place where you study or go for shopping), or is a such destination within the operational
area of at least one {car, bike, scooter}-sharing service?
• Yes,
• No
• I do not know.
Q7. If you think about your daily travel at this time of the year (for work, study, food purchase, etc.), how often
do you use the following transport modes?
(If yes in Q1) Q7.1. Private car as a driver.
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.2. Private car as a passenger.
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.3. Car-sharing (either as a driver or as a passenger).
• Daily

292
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.4. Public Transport (train, intercity, or urban services).
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.5. Motorcycle/scooter.
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.6. Taxi.
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.7. Personal bike
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.8. Bike-sharing
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.9. Scooter-sharing
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q7.10. Walking.

293
• Daily
• 4-6 days a week
• 1-3 days a week
• Once/a few times a month
• Rarely
• Never
Q8. Some activities are listed below. Which transport mode are you most likely to use in such situations? Please
select only one option (the first that comes to mind).
Q8.1. Going to work or school.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.2. Visiting a close relative / friends / relatives / family.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.3. Running an errand in the city center.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.4. Going out for dinner.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi

294
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.5. Taking an excursion in nice weather.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.6. Visiting a shopping center.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.7. Going to smaller shops.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking
• Other
Q8.8. Weekend activities.
• (If yes in Q1) Private car as a driver
• Private car as a passenger
• Car-sharing
• Public transport
• Moto/Scooter
• Taxi
• Personal bike
• Bike-sharing
• Scooter-sharing
• Walking

295
• Other
Q9. In your opinion, which of the following advantages might induce you to use (or use more) {car, bike, scooter}-
sharing? Multiple answers are possible (maximum 3).
Respondents can choose up to 3 options at a time.
• Availability of shared cars near my home/workplace.
• To reduce expenses, such as maintenance and insurance
• To travel more sustainably.
• Increased comfort when traveling.
• The convenience of having a car only when I need it.
• To avoid responsibilities with maintenance and repairs of my own car.
• {Bike-sharing and scooter-sharing questionnaire only: Smooth track without slope}.
If 1, 2, or 3 in Q2
Q10. In your opinion, which of the following weather conditions can make you use the {car, bike, scooter}-sharing
service more than other modes of transportation? Multiple answers are possible (maximum 3).
Respondents can choose up to 3 options at a time.
• Bad weather (e.g., rainy or snowy weather).
• Good weather (e.g., sunny weather).
• Scorching weather.
• Favorable air temperature.
• Freezing weather.
• High humidity level.
• Favorable humidity level.
• High air pollution.
• Low air pollution.
• In winter.
• In spring.
• In summer.
• In autumn.
Q11. In your opinion, which of the following situations might induce you to use (or use more) {car, bike, scooter}-
sharing?
• Travel less than 5 km
• Travel 5 km or more
• Both
Q12. In your opinion, which of the following situations might induce you to use (or use more) {car, bike, scooter}-
sharing?
• Travel less than 30 min
• Travel 30 min or more
• Both
Q13. In your opinion, which of the following situations might induce you to use (or use more) {car, bike, scooter}-
sharing?
• Travel during peak hours
• Travel during off-peak hours
• Both
Q14. In your opinion, which of the following situations might induce you to use (or use more) {car, bike, scooter}-
sharing? (Multiple answers are possible (maximum 3).
Respondents can choose up to 3 options at a time.
• Travel on a weekday morning

296
• Travel on a weekend morning
• Travel on a weekday evening
• Travel on a weekend evening
Q15. In your opinion, which of the following situations might induce you to use (or use more) {car, bike, scooter}-
sharing?
• Travel for leisure (e.g., vising friends or shopping)
• Travel for non-leisure (going to work/school)
• Both
If 1 or 2 in Q2
Q16. The following statements are about your perceptions of {car, bike, scooter}-sharing use. Note that there are
no right or wrong answers for these. However, we are interested in your impressions on this topic. Please, indicate
to what extent you agree or disagree with the following statements.
Q16.1. It is possible for me to use {car, bike, scooter}-sharing for my regular trips.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.2. I am sure I can choose {car, bike, scooter}-sharing for my regular trips during the next week.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.3. The {car, bike, scooter}-sharing service is a useful mode of transport.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.4. {car, bike, scooter}-sharing helps me to accomplish activities that are important to me.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.5. Learning how to use {car, bike, scooter}-sharing was easy for me.
• 1 (Strongly disagree)
• 2
• 3

297
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.6. I find {car, bike, scooter}-sharing easy to use.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q16.7. It is difficult to book a car at the {car, bike, scooter}-sharing website/app.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 1 or 2 in Q2
Q17. Now there are some statements about your social network. To what extent do you agree or disagree with the
following statements?
Q17.1. People who are important to me think I should use {car, bike, scooter}-sharing more often instead of other
modes of transportation.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q17.2. People who are important to me like that I use {car, bike, scooter}-sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q17.3. People who are important to me agree with my use of {car, bike, scooter}-sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 3 in Q2

298
Q18. Now there are some statements about your social network. To what extent do you agree or disagree with the
following statements?
Q18.1. People who are important to me think I should use {car, bike, scooter}-sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q18.2. People who are important to me would like me to use {car, bike, scooter}-sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q18.3. People who are important to me would agree if I used {car, bike, scooter}-sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 1, 2, or 3 in Q2
Q19. The following statements are about {car, bike, scooter}-sharing. Please, indicate the extension of your
opinions.
Q19.1. My support for the implementation of {car, bike, scooter}-sharing in society is
• 1 (Very low)
• 2
• 3
• 4
• 5
• 6
• 7 (Very high)
Q19.2. Overall, my view of {car, bike, scooter}-sharing is
• 1 (Very negative)
• 2
• 3
• 4
• 5
• 6
• 7 (Positive)
If 1 or 2 in Q2
Q20. The following statements are about {car, bike, scooter}-sharing. Please, indicate the extension of your
opinions.

299
Q20.1. Using {car, bike, scooter}-sharing is relatively enjoyable.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q20.2. Using {car, bike, scooter}-sharing is relatively environmentally friendly.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q20.3. The impact of health concerns due to the Covid-19 pandemic has reduced my use of {car, bike, scooter}-
sharing.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 1 or 2 in Q2
Q21. Based on your previous experience with {car, bike, scooter}-sharing, answer the following questions.
Q21.1. I know {car, bike, scooter}-sharing provides good service.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q21.2. I know it is predictable.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q21.3. I know it is trustworthy.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5

300
• 6
• 7 (Strongly agree)
If 3 in Q2
Q22. The following statements are about your perceptions of {car, bike, scooter}-sharing use. Note that there are
no right or wrong answers for these. However, we are interested in your impressions on this topic. Please, indicate
to what extent you agree or disagree with the following statements.
Q22.1. It would be possible for me to use {car, bike, scooter}-sharing for my regular trips.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q22.2. I am sure I can choose the {car, bike, scooter}-sharing for my regular trips during the next week.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q22.3. Using {car, bike, scooter}-sharing services would be a useful mode of transport.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q22.4. Using {car, bike, scooter}-sharing would help me to accomplish activities that are important to me.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q22.5. Learning how to use {car, bike, scooter}-sharing would be easy for me.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q22.6. I would find {car, bike, scooter}-sharing easy to use.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5

301
• 6
• 7 (Strongly agree)
Q22.7. It would be difficult to book a car at the {car, bike, scooter}-sharing website/app.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 3 in Q2
Q23.The following statements are about {car, bike, scooter}-sharing. Please, indicate the extension of your
opinions.
Q23.1. Using {car, bike, scooter}-sharing services would be enjoyable.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q23.2. I think {car, bike, scooter}-sharing services are environmentally friendly.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
If 3 in Q2
Q24. Answer the following questions according to your knowledge of {car, bike, scooter}-sharing.
Q24.1. I think it provides good service.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q24.2. I think it is predictable.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q24.3. I think it is trustworthy.
• 1 (Strongly disagree)
• 2

302
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q25. The following statements are about the environmental impacts of travel. Indicate to what extent you agree
or not.
Q25.1. The urgent need to reduce ecological destruction caused by using cars has been overestimated.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q25.2. I believe that using a car causes many environmental problems.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q26. The following statements are about the environmental impact of your personal daily travels. To which extent
do you agree or disagree with them?
Q26.1. I feel morally obliged to reduce the environmental impact due to my travel patterns.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q26.2. I would feel guilty if I did not reduce the environmental impact of my travel patterns.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)
Q26.3. I would feel good if I traveled more sustainably.
• 1 (Strongly disagree)
• 2
• 3
• 4
• 5
• 6
• 7 (Strongly agree)

303
Q27. Political issues are sometimes referred to on a green environmental scale. Where would you place yourself
on such a green scale?
• 1 (Not green)
• 2
• 3
• 4
• 5
• 6
• 7 (Very green)
Q28. Political issues are sometimes also referred to as "left" and "right". Where would you place your views on
this scale?
• Far to the left
• Left
• Quite left
• Neither to the left nor the right
• Quite right
• Right
• Far to the right
Final questions about yourself (question set C)
Q29. What is your gender?
• Male
• Female
• Other
Q30. In which year were you born?
Select the year.
Q31. What is your marital status?
• Single
• Married or domestic partnership
Q32. What is your business or professional status?
• Entrepreneur/freelancer
• Officer/manager
• Clerk/trade employee
• Worker
• Teacher
• Representative
• Craftsman / trader / operator
• Student
• Housewife
• Retired
• Waiting for first job / never worked
• Unemployed / lost his/her job.
• Other
Q33. What is the highest level of education you have?
• Not completed primary school
• Elementary school
• Upper secondary school or equivalent shorter than 3 years
• Upper secondary school or equivalent 3 years or more
• Post-secondary education, not college, less than 3 years

304
• Post-secondary education, not college 3 years or more
• University less than 3 years
• University 3 years or more
• Degree from postgraduate studies
Q34. Please, select the municipality where you live.
…………
Q35. How many people, including yourself, live in your household?
• 1
• 2
• 3
• 4
• 5 or more
Q36. How many drivers, including yourself, are there in your household?
• 0
• 1
• 2
• More than 2
Q37. Do you have children living in your household?
• Yes
• No
If yes, in Q37
Q38. How old are your children? (You can select more than one option)
Respondents can choose more than one option.
• 0-3 years old
• 4-6 years old
• 7-15 years old
• 16 years old or older
Q39. How many cars are available in your household? (Please also count company cars you have received from
your employer that are authorized for personal use).
• No car
• One car
• Two cars
• Three or more cars
Q40_01. Approximately what is your personal monthly income after taxes?
• Up to 500Euro
• 501Euro - 1000Euro
• 1001Euro - 1500Euro
• 1501Euro - 2000Euro
• 2001Euro - 2500Euro
• 2501Euro - 3000Euro
• 3001Euro - 4000Euro
• 4001Euro - 5000Euro
• 5001Euro - 6000Euro
• € 6001 - € 10,000
• More than 10.001 Euros

305
Q40_02. Approximately what is your monthly household income after taxes? You can answer this question even
if you are unsure of the exact amount.
• Up to 500Euro
• 501Euro - 1000Euro
• 1001Euro - 1500Euro
• 1501Euro - 2000Euro
• 2001Euro - 2500Euro
• 2501Euro - 3000Euro
• 3001Euro - 4000Euro
• 4001Euro - 5000Euro
• 5001Euro - 6000Euro
• € 6001 - € 10,000
• More than 10.001 Euros
Q41. How do you manage your expenses with your current income?
• Very good
• Fairly good
• Neither good nor bad
• Pretty bad
• Very bad
A2.2 Questionnaires for government members and operators of each shared
mobility service (surveys 4 to 6)
This type of survey (4 to 6) is designed for government members and operators of car sharing,
bike-sharing, and scooter-sharing services, and it includes two parts. In the first part, there are
questions related to BWM analysis. In the second part, questions are relevant to the
respondent's opinion about some of the characteristics that might induce people to use (or use
more) {car, bike, scooter}-sharing.
Dear Ms./Mr.
We are conducting a study at Politecnico di Torino. We aim to understand better individuals’ views on the
importance of different characteristics in {car, bike, scooter}-sharing, to know their mobility routines and
daily travels. We assure you that all information you provide will be treated with the utmost confidentiality
and will be completely anonymous. Your participation is a valuable contribution to this study, and we thank
you for your cooperation.
Please answer the following questions (question set A).
B1. There are several trip-related characteristics that could be considered by individuals in selecting {car, bike,
scooter}-sharing to make a trip. These characteristics are listed below.
• Travel time: the time it takes with a given means to travel from origin to destination.
• Travel distance: the distance between origin and destination.
• Departure time: the trip's start time, such as in the morning or evening, on weekends, or on weekdays,
during peak or off-peak hours.
• Trip purpose: the purpose of the trip, such as traveling to work, school, shopping, or meeting a friend.
In your opinion, what is the MOST IMPORTANT, and what is the LEAST IMPORTANT trip-related
characteristic among the above four that could drive individuals’ choice?
Trip-related
characteristics
Select the most important characteristic in the cell
below
Select the least important characteristic in the cell
below
Travel time
Travel distance

306
Departure time
Trip purpose
“4*2=8 radio buttons in the above table are needed to make the selections”.
B2. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other three characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important,
and the other three should be mentioned in the first column according to the template below. 9*3 = 27 radio
buttons should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent
selected in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
Characteristic 2
Characteristic 3
B3. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other two characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other two characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*2 = 18 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Characteristic 3 is more important
than LEAST IMPORTANT
B4. Now, let us examine the relative importance of some {car, bike, scooter}-sharing characteristics. These
characteristics are listed below.
• Cost: expenses for {car, bike, scooter}-sharing usage such as service subscription fees or usage fees.
• Comfort: vehicle characteristics that make you feel comfortable during the trip.
• Safety: The level of safety of the individual during the trip, such as the rate of accidents, harassment,
assault, and theft.
• Service quality: Quality of the {car, bike, scooter}-sharing system and given services.
• Environment-friendly system: a system that is reducing environmental impacts.
• User-friendliness: easy for beginners to learn, easy to use, and provide travel information in the app.
In your opinion, what is the MOST IMPORTANT, and what is the LEAST IMPORTANT {car, bike, scooter}-
sharing characteristic among the above six that could drive individuals’ choice?
{Car, bike, scooter}-sharing
characteristics
Select the most important characteristic
in the cell below
Select the least important characteristic
in the cell below
Travel cost
Travel comfort
Safety
Service quality
Environment-friendly system
User-friendly
“6*2=12 radio buttons in the above table are needed to make the selections
B5. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other five characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other five should be mentioned in the first column according to the template below. 9*5 = 45 radio buttons

307
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Least important)
Characteristic 2
Characteristic 3
Characteristic 4
Characteristic 5
B6. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other four characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other four characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*4 = 36 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Characteristic 3 is more important
than LEAST IMPORTANT
Characteristic 4 is more important
than LEAST IMPORTANT
Characteristic 5 is more important
than LEAST IMPORTANT
B7. Finally, let us consider the following two characteristics related to where shared cars are actually available:
• Service availability: Availability of {car, bike, scooter}-sharing services around shopping malls,
colleges, transportation centers, city centers, and densely populated areas.
• Vehicle availability and accessibility: Availability of the vehicle where I need it, easiness to reach and
access the vehicle, proximity to the location of the parked vehicle from my starting point.
In your opinion, what is the MOST IMPORTANT factor between these two?
• Service availability
• Vehicle availability and accessibility
B8. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than LEAST IMPORTANT?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other one should be mentioned in the first column according to the template below. 9*1 = 9 radio buttons
should appear in the table.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
B9. Now, let us jointly consider trip-related characteristics, {car, bike, scooter}-sharing characteristics, and
availability & accessibility that you separately assessed in the previous questions. In your opinion, which of these
three sets of characteristics is overall the MOST IMPORTANT, and which is the LEAST IMPORTANT when
individuals are considering to choose {car, bike, scooter}-sharing to make a trip?
Characteristics
Select the most important characteristic
in the cell below
Select the least important characteristic
in the cell below

308
Trip-related characteristics (travel time,
travel distance, departure time, trip
purpose)
{car, bike, scooter}-sharing
characteristics
(Cost, comfort, safety, service quality,
environment-friendly system, user-
friendliness)
Availability and accessibility
(Service availability, vehicle availability
and accessibility)
3*2=6 radio buttons in the above table are needed to make the selections.
B10. In the above question, you have chosen MOST_IMPORTANT as the most important set of characteristics.
Could you please rate to which extent you consider MOST_IMPORTANT more important than the other two sets?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other two should be mentioned in the first column according to the template below. 9*2 = 18 radio buttons
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Least important)
Characteristic 2
B11. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other characteristic more important than LEAST_IMPORTANT?
The respondent should see the following table, where the other characteristic, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the below template. 9*1 = 9 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Please answer the following questions to determine your opinion about some of the characteristics that
might induce people to use (or use more) {car, bike, scooter}-sharing (question set D)
Q1. In your opinion, which of the following characteristics might induce people to use (or use more) {car, bike,
scooter}-sharing?
• Short-distance trips (less than 5 km)
• Long-distance trips (beyond 5 km)
• Both
Q2. In your opinion, which of the following characteristics might induce people to use (or use more) {car, bike,
scooter}-sharing?
• Short-time trips (less than 30 min)
• Long-distance trips (beyond 30 min)
• Both
Q3. In your opinion, which of the following characteristics might induce people to use (or use more) {car, bike,
scooter}-sharing?
• During peak hours
• During off-peak hours

309
• Both
Q4. In your opinion, which of the following characteristics might induce people to use (or use more) {car, bike,
scooter}-sharing? (Multiple answers are possible (maximum 3).
Respondents can choose up to 3 options at a time.
• On weekday morning
• On weekend morning
• On weekday evening
• On weekend evening
Q5. In your opinion, which of the following characteristics might induce people to use (or use more) {car, bike,
scooter}-sharing?
• For leisure trips (e.g., vising friends or shopping)
• For non-leisure trips (going to work/school)
• Both
A2.3 Questionnaires for users and non-users of shared mobility services (as
a whole, not for a specific shared mobility service) (survey 7)
This type of survey (survey 7) is designed for users and non-users of shared mobility services
(as a whole, not for a specific shared mobility service), and it includes two parts. In the first
part, there are questions related to BWM analysis. In the second part, questions are relevant to
the respondent's opinions on characteristics affecting car-sharing, bike-sharing, and scooter-
sharing use.
Dear Ms./Mr.
We are conducting a study at Politecnico di Torino. We aim to understand better your views on the
importance of different characteristics in shared mobility and to know your mobility routines and your
daily travels. We assure you that all information you provide will be treated with the utmost confidentiality
and will be completely anonymous. Your participation is a valuable contribution to this study, and we thank
you for your cooperation.
Please read the shared mobility definition first.
Shared mobility definition: shared mobility is a shared vehicle that people can use in many cities and
communities. As a user, you have access to bookable shared vehicles. The vehicles are available 24 hours a
day, 7 days a week, and available through self-service. It is important to note that a trip through shared
mobility is not shared with other users, but it is only the vehicles that are shared with others who use them
at other times.
Please answer the following questions (question set A).
B1. There are several characteristics that could be considered in selecting shared mobility to make a trip. These
characteristics are listed below.
• People’s Safety: The level of safety of the individuals during the trip, such as the rate of accidents,
harassment, assault, and theft.
• Operational speed: the average velocity that a shared mobility system overpasses.
• Accessibility: Ease of access, availability of a shared vehicle, proximity to the location of the parked
shared vehicle.
• User-friendliness: easy for beginners to learn, easy to use, and provide travel information in the app.
• Image: The image of a shared mobility system in the eyes of you.
• Comfort: Vehicle characteristics that make you feel comfortable during the trip
• Cost: Expenses for shared mobility usage, such as service subscription fees or usage fees.
• Possibility of carrying items: Possibility of carrying luggage or bags or shopping items in the shared
vehicle. For instance, people can carry their luggage by shared car, but not by scooter-sharing.

310
In your opinion, what is the MOST IMPORTANT and what is the LEAST IMPORTANT characteristic among
the above eight that could drive your choice?
Characteristics
Select the most important Characteristic
in the cell below
Select the least important Characteristic
in the cell below
People’s safety
Operational speed
Accessibility
User-friendliness
Image
Comfort
Cost
Possibility of carrying items
“8*2=16 radio buttons in the above table are needed to make the selections”.
B2. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other seven
characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important,
and the other seven should be mentioned in the first column according to the template below. 9*7 = 63 radio
buttons should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent
selected in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
Characteristic 2
Characteristic 3
Characteristic 4
Characteristic 5
Characteristic 6
Characteristic 7
B3. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other six characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other six characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*6 = 54 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Characteristic 3 is more important
than LEAST IMPORTANT
Characteristic 4 is more important
than LEAST IMPORTANT
Characteristic 5 is more important
than LEAST IMPORTANT
Characteristic 6 is more important
than LEAST IMPORTANT
Characteristic 7 is more important
than LEAST IMPORTANT
Please answer the following questions to determine your opinion on characteristics affecting car-sharing,
bike-sharing, and scooter-sharing use (question set E).
Q1. How safe do you feel on car-sharing trips?
• 1 (Very unsafe)
• 2

311
• 3
• 4
• 5
• 6
• 7 (Very safe)
Q2. How safe do you feel on bike-sharing trips?
• 1 (Very unsafe)
• 2
• 3
• 4
• 5
• 6
• 7 (Very safe)
Q3. How safe do you feel on scooter-sharing trips?
• 1 (Very unsafe)
• 2
• 3
• 4
• 5
• 6
• 7 (Very safe)
Q4. How would you rate the travel speed of a car-sharing service?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q5. How would you rate the travel speed of a bike-sharing service?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q6. How would you rate the travel speed of the scooter-sharing service?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q7. How easy or difficult is it to access car-sharing?
• 1 (Very difficult)
• 2
• 3

312
• 4
• 5
• 6
• 7 (Very easy)
Q8. How easy or difficult is it to access bike-sharing?
• 1 (Very difficult)
• 2
• 3
• 4
• 5
• 6
• 7 (Very easy)
Q9. How easy or difficult is it to access scooter-sharing?
• 1 (Very difficult)
• 2
• 3
• 4
• 5
• 6
• 7 (Very easy)
Q10. How would you rate the user-friendliness of car-sharing services?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q11. How would you rate the user-friendliness of bike-sharing services?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q12. How would you rate the user-friendliness of the scooter-sharing services?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q13. How would you rate car-sharing service overall?
• 1 (Very poor)
• 2
• 3
• 4

313
• 5
• 6
• 7 (Very good)
Q14. How would you rate bike-sharing service overall?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q15. How would you rate the scooter-sharing service overall?
• 1 (Very poor)
• 2
• 3
• 4
• 5
• 6
• 7 (Very good)
Q16. How comfortable do you feel on car-sharing trips?
• 1 (Very uncomfortable)
• 2
• 3
• 4
• 5
• 6
• 7 (Very comfortable)
Q17. How comfortable do you feel on bike-sharing trips?
• 1 (Very uncomfortable)
• 2
• 3
• 4
• 5
• 6
• 7 (Very comfortable)
Q18. How comfortable do you feel on scooter-sharing trips?
• 1 (Very uncomfortable)
• 2
• 3
• 4
• 5
• 6
• 7 (Very comfortable)
Q19. How would you rate the usage or membership fees of car-sharing services?
• 1 (Very expensive)
• 2
• 3
• 4
• 5

314
• 6
• 7 (Very cheap)
Q20. How would you rate the usage or membership fees of bike-sharing services?
• 1 (Very expensive)
• 2
• 3
• 4
• 5
• 6
• 7 (Very cheap)
Q21. How would you rate the usage or membership fees of scooter-sharing services?
• 1 (Very expensive)
• 2
• 3
• 4
• 5
• 6
• 7 (Very cheap)
Q22. Is it difficult or easy to carry your belongings when using car-sharing?
• 1 (Very difficult)
• 2
• 3
• 4
• 5
• 6
• 7 (Very easy)
Q23. Is it difficult or easy to carry your belongings when using bike-sharing?
• 1 (Very difficult)
• 2
• 3
• 4
• 5
• 6
• 7 (Very easy)
Q24. Is it difficult or easy to carry your belongings when using scooter-sharing?
• 1 (Very difficult)
• 2
• 3
• 4
• 5
• 6
• 7 (Very easy)
A2.4 Questionnaire for government members about shared mobility services
(as a whole, not for a specific shared mobility service) (surveys 8)
This type of survey (survey 8) is designed for government members and is about shared
mobility services (as a whole, not for a specific shared mobility service). In this survey, there
are questions related to BWM analysis.

315
Dear Ms./Mr.
We are conducting a study at Politecnico di Torino. We aim to understand better individuals’ views on the
importance of different characteristics in shared mobility, to know their mobility routines and their daily
travels. We assure you that all information you provide will be treated with the utmost confidentiality and
will be completely anonymous. Your participation is a valuable contribution to this study, and we thank
you for your cooperation.
Please briefly state your role in your Administration. [Open Question]
Question set A:
Suppose, as a government member, you want to decide on a new shared mobility system to be set up in Turin,
Italy. The following characteristics are considered to select the most appropriate system among the following
three: car-sharing, bike-sharing, and scooter-sharing. You could make your decision based on the following
characteristics.
• The number of trips per vehicle per day: it gives insight into the efficiency of the vehicle that shows
the efficiency of the service.
• Greenhouse gases (GHGs): the amount of greenhouse gas emissions by a shared mobility system.
• Parking issues: illegal parking of shared vehicles like parking in inappropriate places.
• Emission of pollutants: pollutants emitted by a shared vehicle.
• Integration of the shared mobility service with public transport: Complementarity of a shared
vehicle for public transport. Their integration can increase urban mobility.
• Vehicle fee: the fee that a shared mobility operator may pay to the municipality. For example, car-sharing
operators pay a fee to the municipality, which allows their shared cars to go to city centers or places
where traffic is restricted.
Please Answer the following questions.
Do you think something is missing from the list above? The above characteristics are important criteria that make
it possible to compare shared mobility modes (car-sharing, bike-sharing, scooter-sharing). What do you think
about this list? Are there any unmentioned or unclear criteria? Do you have anything to add?
B1. In your opinion, what is the MOST IMPORTANT, and what is the LEAST IMPORTANT characteristic
among the above six that could drive your choice?
Characteristics
Select the most important characteristic
in the cell below
Select the least important characteristic
in the cell below
The number of trips per vehicle per day
Greenhouse gases (GHGs)
Parking issues
Emission of pollutants
Integration of the shared mobility service
with public transport
Vehicle fee
“6*2=12 radio buttons in the above table are needed to make the selections”.
B2. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other five characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other five should be mentioned in the first column according to the template below. 9*5 = 45 radio buttons
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
Characteristic 2
Characteristic 3
Characteristic 4
Characteristic 5

316
B3. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other four characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other four characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*4 = 36 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Characteristic 3 is more important
than LEAST IMPORTANT
Characteristic 4 is more important
than LEAST IMPORTANT
Characteristic 5 is more important
than LEAST IMPORTANT
A2.5 Questionnaire for operators of shared mobility services (as a whole, not
for a specific shared mobility service) (survey 9)
This type of survey (survey 9) is designed for operators of shared mobility services (as a whole,
not for a specific shared mobility service). In this survey, there are questions related to BWM
analysis.
Dear Ms./Mr.
We are conducting a study at Politecnico di Torino. We aim to understand better individuals’ views on the
importance of different characteristics in shared mobility, to know their mobility routines and their daily
travels. We assure you that all information you provide will be treated with the utmost confidentiality and
will be completely anonymous. Your participation is a valuable contribution to this study, and we thank
you for your cooperation.
Which kind of shared mobility service is offered by your company? (You can choose more than one option)
• Free-floating car-sharing
• Station-based car-sharing
• Free-floating bike-sharing
• Station-based bike-sharing
• Scooter-sharing
Please briefly state your role in your company. [Open Question]
Question set A:
As a shared mobility operator, suppose you plan to run your own shared mobility system in a city. The following
characteristics are already known as important elements for system implementation.
• Vehicle utilization rate (%): total time (minutes) that all shared vehicles are used each day divided by
the time they can potentially be used per day in 24 hours, which shows the efficiency of the service.
• Usage fees (membership fees) (€): Operators can experience higher revenue with higher usage fees
(membership fees), and it affects earnings. Suppose you are free to set the price of your services.
• The average number of trips per vehicle per day: it gives insight into the efficiency of the vehicle that
shows the efficiency of the service.
• Operational speed (Km/h): the average velocity a shared mobility system overpasses.

317
• The Lifespan of vehicle (year): system lifespan is measured in years and is indicated by the lifespan of
vehicles.
Please Answer the following questions.
Do you think something is missing from the list above? The above factors are important criteria that make it
possible to compare shared mobility modes (car-sharing, bike-sharing, scooter-sharing). What do you think about
this list? Are there any unmentioned or unclear criteria? Do you have anything to add?
B1. In your opinion, what is the MOST IMPORTANT and what is the LEAST IMPORTANT characteristic among
the above five that could drive your choice?
Characteristics
Select the most important Characteristic
in the cell below
Select the least important Characteristic
in the cell below
Utilization rate
Usage fees
The number of trips per vehicle per day
Operational speed
The life span of the vehicle
5*2=10 radio buttons in the above table are needed to make the selections”.
B2. In the above question, you have chosen MOST_IMPORTANT as the most important characteristic. Could you
please rate to which extent you consider MOST_IMPORTANT more important than the other four characteristics?
The respondent should see the following table, where the characteristic which is selected as the most important
and the other four should be mentioned in the first column according to the template below. 9*4 = 36 radio buttons
should appear in the table. Characteristic 1 is always "the least important characteristic" the respondent selected
in the previous step.
To which extent MOST
IMPORTANT is more
important than…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
LEAST IMPORTANT
Characteristic 2
Characteristic 3
Characteristic 4
B3. Also, you have chosen LEAST_IMPORTANT as the least important characteristic. Could you please rate to
which extent you consider the other three characteristics more important than LEAST_IMPORTANT?
The respondent should see the following table. The other three characteristics, neither MOST IMPORTANT nor
LEAST IMPORTANT, should be mentioned in the first column according to the template below. 9*3 = 27 radio
buttons should appear in the table.
To which extent…
Equal
importance
1
2
3
4
5
6
7
8
Extremely
more
important
9
Characteristic 2 is more important
than LEAST IMPORTANT
Characteristic 3 is more important
than LEAST IMPORTANT
Characteristic 4 is more important
than LEAST IMPORTANT

318
Appendix 3
Appendix 3: Codebook
Codebooks contribute to describing the data collection's contents, structure, and layout. In this
study, there are nine different surveys whose questions are reported in appendix 2, leading to
nine different codebooks. Since car-sharing, bike-sharing, and scooter-sharing codebooks are
similar, one general codebook is provided. In this regard, instead of specifying the type of
service in the general codebook, it is written as "{car, bike, scooter}-sharing}," meaning that
this general codebook can be used for each of these three shared mobility services. Besides,
three different codebooks are provided separately for users/non-users, government members,
and operators of shared mobility services (as a whole). It is important to note that since the
surveys were conducted in Italian, the codebooks are also in Italian and are presented in A3.1
to A3.5. Also, in this section, the job positions of government members and operators are listed
according to the type of shared transportation service in section A3.6. This list has been
translated into English because this list is the answers of people to the survey questions. In this
study, the individuals' responses to the survey questions have been translated into English.
A3.1 The codebook for users and non-users of {car, bike, scooter}-
sharing (general codebook) (surveys 1 to 3)
This codebook is designed for users and non-users of car-sharing, bike-sharing, and scooter-
sharing services. This type of general codebook is presented as follows.
B1
Valore
Etichetta
B1. Ci sono diverse caratteristiche relative agli spostamenti che potrebbero essere considerate nella scelta del
{car, bike, scooter}-sharing per effettuare uno spostamento.
B1_01
Valore

319
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio
e qual è quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Tempo di percorrenza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_02
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio
e qual è quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Distanza da percorrere
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_03
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio
e qual è quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Orario di partenza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_04
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio
e qual è quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Scopo del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B2_01
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare fino a che punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_02

320
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare fino a che punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_03
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare fino a che punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_01
Valore
Etichetta
B3. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_02
Valore

321
Etichetta
B3. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B4
Valore
Etichetta
B4. Ora, esaminiamo l'importanza relativa di alcune caratteristiche del {car, bike, scooter}-sharing.
B4_01
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Costo del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_02
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Comfort del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_03
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Sicurezza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_04
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Qualità del servizio

322
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_05
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Sistema rispettoso dell'ambiente
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_06
Valore
Etichetta
B4. Secondo lei, tra le sei caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero
influenzare la sua scelta, qual è la caratteristica PIÙ IMPORTANTE e qual è quella MENO IMPORTANTE?
Facilità di utilizzo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B5_01
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare in che misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_02
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare in che misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7

323
8
8
9
9 Estremamentepiu' importante
B5_03
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare in che misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_04
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare in che misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_05
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE come caratteristica più importante. Potrebbe
per favore valutare in che misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque
caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8

324
9
9 Estremamentepiu' importante
B6_01
Valore
Etichetta
B6. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B6_02
Valore
Etichetta
B6. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B6_03
Valore
Etichetta
B6. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante

325
B6_04
Valore
Etichetta
B6. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera le altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B7
Valore
Etichetta
B7. Infine, consideriamo le seguenti due caratteristiche relative al luogo in cui le auto condivise sono
effettivamente disponibili. Secondo lei, qual tra questi due è il fattore PIÙ IMPORTANTE?
1
Disponibilita' del servizio
2
Disponibilita' e accessibilita' del veicolo
B8
Valore
Etichetta
B8. Alla domanda precedente, lei ha scelto PIU’_IMPORTANTE quale caratteristica più importante. Potrebbe
per favore valutare fino a che punto considera questa caratteristica PIÙ IMPORTANTE di quella MENO
IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B9
Valore
Etichetta
B9. Ora, consideriamo insieme le caratteristiche relative allo spostamento, le caratteristiche del car sharing e la
disponibilità e l'accessibilità che ha valutato separatamente nelle domande precedenti.
B9_01
Valore

326
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale
è il MENO IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
Caratteristiche relative al viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_02
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale
è il MENO IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_03
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale
è il MENO IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
Caratteristiche del {car, bike, scooter}-sharing
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_04
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale
è il MENO IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_05
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale
è il MENO IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
Disponibilità ed accessibilità
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B10_01
Valore
Etichetta
B10. Alla domanda precedente, lei ha scelto il gruppo PIU’_IMPORTANTE come gruppo di caratteristiche più
importanti. Potrebbe per favore valutare fino a che punto considera il gruppo PIU’_IMPORTANTE più
importante degli altri due gruppi?
1
1 Uguale importanza
2
2
3
3

327
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B10_02
Valore
Etichetta
B10. Alla domanda precedente, lei ha scelto il gruppo PIU’_IMPORTANTE come gruppo di caratteristiche più
importanti. Potrebbe per favore valutare fino a che punto considera il gruppo PIU’_IMPORTANTE più
importante degli altri due gruppi?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B11
Valore
Etichetta
B11. Inoltre, lei ha scelto MENO_IMPORTANTE come caratteristica meno importante. Potrebbe per favore
valutare in che misura considera l'altra caratteristica più importante di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
INTRO
Valore
Etichetta
Le seguenti domande riguardano le sue abitudini quotidiane e opinioni di viaggio.
1
Explanation
D1
Valore
Etichetta
D1. Ha una patente di guida?
1
Si'
2
No

328
D2
Valore
Etichetta
D2. Ha qualche esperienza di servizi di {car, bike, scooter}-sharing?
1
Si', attualmente uso i servizi di {car, bike, scooter}-sharing
2
Si', ho usato il {car, bike, scooter}-sharing in passato, ma non lo uso più
3
No, non ho mai usato il {car, bike, scooter}-sharing, ma so cos'e'
4
Non ho familiarita' con il concetto di {car, bike, scooter}-sharing
Users
Valore
Etichetta
Utilizzo
1
users
2
non-users
D3
Valore
Etichetta
D3. Quanto conosce il {car, bike, scooter}-sharing? (Termini di adesione, come prenotare, livelli di prezzo, ecc.)
1
1 (Poco)
2
2
3
3
4
4
5
5 (Molto)
D4_01
Valore
Etichetta
(Scelta 1)D4. Quali servizi di {car, bike, scooter}-sharing utilizza?
1
Enjoy
2
Car2go (Share Now)
3
BlueTorino
4
Helbiz An
5
Circ
6
Lime
7
Wind
8
Link
9
Vo i
10
Altro (specificare)
D4_02
Valore
Etichetta
(Scelta 2)D4. Quali servizi di {car, bike, scooter}-sharing utilizza?
1
Enjoy
2
Car2go (Share Now)
3
BlueTorino
4
Helbiz An
5
Circ

329
6
Lime
7
Wind
8
Link
9
Vo i
10
Altro (specificare)
D4_03
Valore
Etichetta
(Scelta 3)D4. Quali servizi di {car, bike, scooter}-sharing utilizza?
1
Enjoy
2
Car2go (Share Now)
3
BlueTorino
4
Helbiz An
5
Circ
6
Lime
7
Wind
8
Link
9
Vo i
10
Altro (specificare)
D4_04
Valore
Etichetta
(Scelta 4)D4. Quali servizi di {car, bike, scooter}-sharing utilizza?
1
Enjoy
2
Car2go (Share Now)
3
BlueTorino
4
Helbiz An
5
Circ
6
Lime
7
Wind
8
Link
9
Vo i
10
Altro (specificare)
D4_text
Valore
Etichetta
D4. Quali servizi di {car, bike, scooter}-sharing utilizza?
Valori
validi
MIMOTO
Share
now
D5
Valore
Etichetta
D5. Ci sono punti di ritiro del {car, bike, scooter}-sharing vicino a casa sua, o la sua casa si trova in un'area
operativa di almeno un servizio di {car, bike, scooter}-sharing?
1
Si'
2
No

330
3
Non lo so
D6
Valore
Etichetta
D6. Ci sono punti di ritiro del car sharing vicino alla destinazione più frequente dei suoi spostamenti (ad esempio,
il posto di lavoro, il luogo dove studia o va a fare shopping), o tale destinazione si trova all'interno dell'area
operativa di almeno un
1
Si'
2
No
3
Non lo so
D7_01
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.1. Auto privata come autista
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_02
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.2. Auto privata come passeggero
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_03
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.3. {car, bike, scooter}-sharing (sia come autista che
come passeggero)
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_04

331
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.4. Trasporto pubblico (treno, intercity o servizi urbani)
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_05
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.5. Moto/scooter
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_06
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.6. Taxi
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_07
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.7. Bicicletta personale
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_08
Valore

332
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.8. Bike-sharing
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_09
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.9. Monopattino in condivisione
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D7_10
Valore
Etichetta
D7. Se pensa ai suoi spostamenti quotidiani in questo periodo dell'anno (per lavoro, studio, acquisto di cibo, ecc.),
quanto spesso usa le seguenti modalita' di trasporto? : D7.10. Camminare a piedi
1
Tutti i giorni
2
4-6 giorni a settimana
3
1-3 giorni a settimana
4
Una volta/alcune volte al mese
5
Raramente
6
Mai
D8_01
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.1. Andare al lavoro o a
scuola
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione

333
10
A piedi
11
Altro
D8_02
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.2. Visitare un parente
stretto/amici/altri pare
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_03
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.3. Fare una commissione
in centro citta'
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_04
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.4. Andare fuori a cena
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico

334
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_05
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.5. Fare un'escursione con
il bel tempo.
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_06
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.6. Andare ad un centro
commerciale.
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_07
Valore

335
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.7. Andare in negozi piu'
piccoli.
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D8_08
Valore
Etichetta
D8. Qui di seguito sono elencate alcune attivita'. Quale modalita' di trasporto e' piu' probabile che lei usi in queste
situazioni? Per favore, selezioni solo un'opzione (la prima che le viene in mente).: D8.8. Attivita' nel fine
settimana.
1
Auto privata come autista
2
Auto privata come passeggero
3
{car, bike, scooter}-sharing(sia come autista che come passeggero)
4
Trasporto pubblico
5
Moto/Scooter
6
Taxi
7
Bicicletta personale
8
Bike-sharing
9
Monopattino in condivisione
10
A piedi
11
Altro
D9_01
Valore
Etichetta
(Scelta 1)D9. Secondo lei, quali dei seguenti vantaggi potrebbero indurla ad utilizzare (o usare maggiormente) il
{car, bike, scooter}-sharing?
1
Disponibilita' di auto condivise vicino alla mia casa/luogo di lavoro
2
Per ridurre le spese, quali la manutenzione e l'assicurazione
3
Per viaggiare in modo piu' sostenibile.
4
Maggiore comodita' quando si viaggia.
5
La comodita' di avere una macchina solo quando ne ho bisogno.
6
Evitare le responsabilita' della manutenzione e delle riparazioni della mia auto
D9_02
Valore
Etichetta
(Scelta 2)D9. Secondo lei, quali dei seguenti vantaggi potrebbero indurla ad utilizzare (o usare maggiormente) il
{car, bike, scooter}-sharing?

336
1
Disponibilita' di auto condivise vicino alla mia casa/luogo di lavoro
2
Per ridurre le spese, quali la manutenzione e l'assicurazione
3
Per viaggiare in modo piu' sostenibile.
4
Maggiore comodita' quando si viaggia.
5
La comodita' di avere una macchina solo quando ne ho bisogno.
6
Evitare le responsabilita' della manutenzione e delle riparazioni della mia auto
D9_03
Valore
Etichetta
(Scelta 3)D9. Secondo lei, quali dei seguenti vantaggi potrebbero indurla ad utilizzare (o usare maggiormente) il
{car, bike, scooter}-sharing?
1
Disponibilita' di auto condivise vicino alla mia casa/luogo di lavoro
2
Per ridurre le spese, quali la manutenzione e l'assicurazione
3
Per viaggiare in modo piu' sostenibile.
4
Maggiore comodita' quando si viaggia.
5
La comodita' di avere una macchina solo quando ne ho bisogno.
6
Evitare le responsabilita' della manutenzione e delle riparazioni della mia auto
D10_01
Valore
Etichetta
(Scelta 1)D10. Secondo lei, quali delle seguenti condizioni meteorologiche possono farle utilizzare il servizio di
{car, bike, scooter}-sharing più di altri mezzi di traspoto?
1
Cattivo tempo (ad esempio, pioggia o neve).
2
Bel tempo (ad esempio, tempo soleggiato).
3
Tempo torrido.
4
Temperatura dell'aria favorevole.
5
Tempo gelido.
6
Alto livello di umidita'.
7
Livello di umidita' favorevole.
8
Alto inquinamento dell'aria.
9
Basso inquinamento dell'aria.
10
In inverno.
11
In primavera.
12
In estate.
13
In autunno.
D10_02
Valore
Etichetta
(Scelta 2)D10. Secondo lei, quali delle seguenti condizioni meteorologiche possono farle utilizzare il servizio di
{car, bike, scooter}-sharing più di altri mezzi di traspoto?
1
Cattivo tempo (ad esempio, pioggia o neve).
2
Bel tempo (ad esempio, tempo soleggiato).
3
Tempo torrido.
4
Temperatura dell'aria favorevole.
5
Tempo gelido.

337
6
Alto livello di umidita'.
7
Livello di umidita' favorevole.
8
Alto inquinamento dell'aria.
9
Basso inquinamento dell'aria.
10
In inverno.
11
In primavera.
12
In estate.
13
In autunno.
D10_03
Valore
Etichetta
(Scelta 3)D10. Secondo lei, quali delle seguenti condizioni meteorologiche possono farle utilizzare il servizio di
{car, bike, scooter}-sharing più di altri mezzi di traspoto?
1
Cattivo tempo (ad esempio, pioggia o neve).
2
Bel tempo (ad esempio, tempo soleggiato).
3
Tempo torrido.
4
Temperatura dell'aria favorevole.
5
Tempo gelido.
6
Alto livello di umidita'.
7
Livello di umidita' favorevole.
8
Alto inquinamento dell'aria.
9
Basso inquinamento dell'aria.
10
In inverno.
11
In primavera.
12
In estate.
13
In autunno.
D11
Valore
Etichetta
D11. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Uno spostamento inferiore ai 5 km
2
Uno spostamento di 5 km o piu'
3
Entrambi
D12
Valore
Etichetta
D12. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Un tempo di viaggio inferiore ai 30 min
2
Un tempo di viaggio di 30 min o piu'
3
Entrambi
D13
Valore
Etichetta
D13. Secondo lei, quale delle seguenti situazioni potrebbe indurla a utilizzare (o utilizzare maggiormente) il {car,
bike, scooter}-sharing?
1
Viaggiare durante le ore di punta

338
2
Viaggiare durante le ore non di punta
3
Entrambi
D14_01
Valore
Etichetta
(Scelta 1)D14. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Viaggiare la mattina dei giorni feriali
2
Viaggiare la mattina del fine settimana
3
Viaggiare la sera dei giorni feriali
4
Viaggiare la sera del fine settimana
D14_02
Valore
Etichetta
(Scelta 2)D14. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Viaggiare la mattina dei giorni feriali
2
Viaggiare la mattina del fine settimana
3
Viaggiare la sera dei giorni feriali
4
Viaggiare la sera del fine settimana
D14_03
Valore
Etichetta
(Scelta 3)D14. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Viaggiare la mattina dei giorni feriali
2
Viaggiare la mattina del fine settimana
3
Viaggiare la sera dei giorni feriali
4
Viaggiare la sera del fine settimana
D15
Valore
Etichetta
D15. Secondo lei, quale delle seguenti situazioni potrebbe indurla ad utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Viaggi di piacere (per esempio, andare a trovare gli amici o fare shopping)
2
Viaggi non di piacere (andare al lavoro o a scuola)
3
Entrambi
D16
Valore
Etichetta
Le seguenti affermazioni riguardano la sua percezione dell'utilizzo del {car, bike, scooter}-sharing. Non ci sono
risposte giuste o sbagliate per queste affermazioni. Siamo interessati al suo punto di vista su questo argomento.
Per favore, indichi in che misura lei è d'ac
D16_01
Valore
Etichetta
D16. È possibile che io utilizzi il {car, bike, scooter}-sharing per i miei viaggi abituali.
1
1 (Fortemente in disaccordo)

339
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_02
Valore
Etichetta
D16. Sono sicuro di poter scegliere il {car, bike, scooter}-sharing per i miei viaggi abituali durante la prossima
settimana.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_03
Valore
Etichetta
D16. Il servizio di {car, bike, scooter}-sharing è un utile mezzo di trasporto.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_04
Valore
Etichetta
D16. Il {car, bike, scooter}-sharing mi aiuta a realizzare attività che sono importanti per me.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_05
Valore
Etichetta
D16. Imparare ad usare il {car, bike, scooter}-sharing è stato facile per me.
1
1 (Fortemente in disaccordo)
2
2

340
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_06
Valore
Etichetta
D16. Trovo il {car, bike, scooter}-sharing facile da usare.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D16_07
Valore
Etichetta
D16. E' difficile prenotare un’auto sul sito web/app del car sharing.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D17_01
Valore
Etichetta
D17. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D17.1. Le persone che sono importanti per me pensano che dovrei usare più spesso il
{car, bike, scooter}-sharing invece di altri mezzi
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D17_02
Valore
Etichetta
D17. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D17.2. Alle persone importanti per me piace che io usi il {car, bike, scooter}-sharing.
1
1 (Fortemente in disaccordo)

341
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D17_03
Valore
Etichetta
D17. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D17.3. Le persone importanti per me sono d'accordo con il mio uso del {car, bike,
scooter}-sharing.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D18_01
Valore
Etichetta
D18. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D18.1. Le persone importanti per me pensano che dovrei usare il {car, bike, scooter}-
sharing.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D18_02
Valore
Etichetta
D18. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D18.2. Le persone importanti per me vorrebbero che io usassi il {car, bike, scooter}-
sharing.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)

342
D18_03
Valore
Etichetta
D18. Quelle che seguono sono alcune affermazioni sui social network. In che misura e' d'accordo o in disaccordo
con queste affermazioni?: D18.3. Le persone importanti per me sarebbero d'accordo se usassi il {car, bike,
scooter}-sharing.
1
1 (Fortemente in disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7 (Fortemente d'accordo)
D19
Valore
Etichetta
D19. Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi in che misura
corrispondono alle sue opinioni.
1
Explanation
D19.1
Valore
Etichetta
D19.1. Il mio sostegno all'attuazione del {car, bike, scooter}-sharing nella società è
1
1 (Molto basso)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto alto)
D19.2
Valore
Etichetta
D19.2. Nel complesso, la mia opinione sul {car, bike, scooter}-sharing è
1
1 (Molto negativa)
2
2
3
3
4
4
5
5
6
6
7
7 (Positiva)
D20_01
Valore
Etichetta
D20. Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi in che misura
corrispondono alle sue opinioni. Usare il {car, bike, scooter}-sharing è relativamente piacevole.
1
1(Fortementein disaccordo)
2
2

343
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D20_02
Valore
Etichetta
D20. Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi in che misura
corrispondono alle sue opinioni. L’utilizzo del {car, bike, scooter}-sharing è relativamente rispettoso
dell'ambiente.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D20_03
Valore
Etichetta
D20.Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi in che misura
corrispondono alle sue opinioni. L'impatto delle preoccupazioni sanitarie dovute alla pandemia di Covid-19 ha
ridotto il mio uso del {car, bike, scooter}-sharing.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D21_01
Valore
Etichetta
D21. In base alla sua precedente esperienza con il {car, bike, scooter}-sharing, risponda alle seguenti domande.
So che il {car, bike, scooter}-sharing fornisce un buon servizio.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D21_02
Valore

344
Etichetta
D21. In base alla sua precedente esperienza con il {car, bike, scooter}-sharing, risponda alle seguenti domande.
So che è prevedibile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D21_03
Valore
Etichetta
D21. In base alla sua precedente esperienza con il {car, bike, scooter}-sharing, risponda alle seguenti domande.
So che è affidabile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D22
Valore
Etichetta
D22. Le seguenti affermazioni riguardano la sua percezione dell'uso del {car, bike, scooter}-sharing. Non ci sono
risposte giuste o sbagliate per queste affermazioni. Siamo interessati al suo punto di vista su questo argomento.
Per favore, indichi in che misura è d'accord
D22_01
Valore
Etichetta
D22. Sarebbe possibile per me utilizzare il {car, bike, scooter}-sharing per i miei spostamenti abituali.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D22_02
Valore
Etichetta
D22. Sono sicuro di poter scegliere il {car, bike, scooter}-sharing per i miei spostamenti abituali durante la
prossima settimana.
1
1(Fortementein disaccordo)
2
2
3
3
4
4

345
5
5
6
6
7
7(Fortemented'accordo)
D22_03
Valore
Etichetta
D22. Usare i servizi di {car, bike, scooter}-sharing sarebbe un modo di trasporto utile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D22_04
Valore
Etichetta
D22. Usare il {car, bike, scooter}-sharing mi aiuterebbe a realizzare attività che sono importanti per me.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D22_05
Valore
Etichetta
D22. Imparare ad usare il {car, bike, scooter}-sharing sarebbe facile per me.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D22_06
Valore
Etichetta
D22. Troverei il {car, bike, scooter}-sharing facile da usare.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6

346
7
7(Fortemented'accordo)
D22_07
Valore
Etichetta
D22. Sarebbe difficile prenotare un’auto sul sito web/app del car sharing.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D23_01
Valore
Etichetta
D23.Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi fino a che punto è
d’accordo con esse. Usare i servizi di {car, bike, scooter}-sharing sarebbe piacevole.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D23_02
Valore
Etichetta
D23.Le seguenti affermazioni riguardano il {car, bike, scooter}-sharing. Per favore, indichi fino a che punto è
d’accordo con esse. Penso che i servizi di {car, bike, scooter}-sharing siano rispettosi dell'ambiente.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D24_01
Valore
Etichetta
D24. Risponda alle seguenti domande in base alla sua conoscenza del {car, bike, scooter}-sharing. Penso che
fornisca un buon servizio.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5

347
6
6
7
7(Fortemented'accordo)
D24_02
Valore
Etichetta
D24. Risponda alle seguenti domande in base alla sua conoscenza del {car, bike, scooter}-sharing. Penso che sia
prevedibile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D24_03
Valore
Etichetta
D24. Risponda alle seguenti domande in base alla sua conoscenza del {car, bike, scooter}-sharing. Penso che sia
affidabile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D25_01
Valore
Etichetta
D25. Le seguenti affermazioni riguardano l'impatto ambientale degli spostamenti. Indichi in che misura e'
d'accordo o in disaccordo. : D25.1. L'urgente necessita' di ridurre la distruzione ecologica causata dall'uso
dell'automobile e' stata sopravvalutata
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D25_02
Valore
Etichetta
D25. Le seguenti affermazioni riguardano l'impatto ambientale degli spostamenti. Indichi in che misura e'
d'accordo o in disaccordo. : D25.2. Credo che l'uso dell'auto causi molti problemi ambientali.
1
1(Fortementein disaccordo)
2
2
3
3

348
4
4
5
5
6
6
7
7(Fortemented'accordo)
D26_01
Valore
Etichetta
D26. Le seguenti affermazioni riguardano l'impatto ambientale dei suoi spostamenti personali quotidiani. In che
misura e' d'accordo o in disaccordo con esse?: D26.1. Mi sento moralmente obbligato a ridurre l'impatto
ambientale dovuto alle mie abitudini di
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D26_02
Valore
Etichetta
D26. Le seguenti affermazioni riguardano l'impatto ambientale dei suoi spostamenti personali quotidiani. In che
misura e' d'accordo o in disaccordo con esse?: D26.2. Mi sentirei in colpa se non riducessi l'impatto ambientale
delle mie abitudini di viaggio
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D26_03
Valore
Etichetta
D26. Le seguenti affermazioni riguardano l'impatto ambientale dei suoi spostamenti personali quotidiani. In che
misura e' d'accordo o in disaccordo con esse?: D26.3. Mi sentirei bene se viaggiassi in modo piu' sostenibile.
1
1(Fortementein disaccordo)
2
2
3
3
4
4
5
5
6
6
7
7(Fortemented'accordo)
D27
Valore

349
Etichetta
D27. Le questioni politiche sono a volte misurate su una scala ambientale verde. Dove si collocherebbe lei su
questa scala verde?
1
1 (Non verde)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto verde)
D28
Valore
Etichetta
D28. Le questioni politiche sono talvolta indicate come di 'sinistra' e di 'destra'. Dove collocherebbe le sue
opinioni su questa scala?
1
Molto a sinistra
2
A sinistra
3
Abbastanza a sinistra
4
Ne' a sinistra ne' a destra
5
Abbastanza a destra
6
A destra
7
Molto a destra
D29
Valore
Etichetta
D29. Di che genere e' lei?
1
Maschio
2
Femmina
3
Altro
D30
Valore
Etichetta
D30. In quale anno e' nato?
Valori
validi
1934
1938
1942
1944
1945
1948
1949
1950
1952
1953
1954
1955
1956
1957
1958

350
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1978
1979
1980
1981
1982
1983
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
D31
Valore
Etichetta
D31. Qual e' il suo stato civile?
1
Celibe/nubile
2
Sposato o convivente

351
D32
Valore
Etichetta
D32. Qual e' il suo status commerciale o professionale?
1
Imprenditore/libero professionista
2
Funzionario/dirigente
3
Impiegato/ operaio specializzato
4
Operaio
5
Insegnante
6
Rappresentante
7
Artigiano / commerciante / operatore
8
Studente
9
Casalinga
10
In pensione
11
In attesa del primo lavoro / mai lavorato
12
Disoccupato / ha perso il lavoro
13
Altro
D33
Valore
Etichetta
D33. Qual e' il livello di istruzione piu' alto che ha conseguito?
1
Non ha completato la scuola elementare
2
Scuola elementare
3
Scuola secondaria superiore o equivalente inferiore ai 3 anni
4
Scuola secondaria superiore o equivalente 3 anni o piu'
5
Istruzione post-secondaria, non universitaria inferiore ai 3 anni
6
Istruzione post-secondaria, non universitaria 3 anni o piu'
7
Universita' inferiore ai 3 anni
8
Universita' 3 anni o piu'
9
Diploma da studi post-laurea
D35
Valore
Etichetta
D35. Quante persone, compreso lei, vivono nel suo nucleo familiare?
1
1
2
2
3
3
4
4
5
5 o piu'
D36
Valore
Etichetta
D36. Quanti guidatori, incluso lei, ci sono nella sua famiglia?
0
0
1
1
2
2
3
Piu' di 2

352
D37
Valore
Etichetta
D37. Ha figli conviventi in famiglia?
1
Si'
2
No
D38_01
Valore
Etichetta
(Scelta 1)D38. Quanti anni hanno i suoi figli? (Puo' selezionare piu' di un'opzione) Gli intervistati possono
scegliere piu' di un'opzione.
1
0-3 anni
2
4-6 anni
3
7-15 anni
4
16 anni o piu'
D38_02
Valore
Etichetta
(Scelta 2)D38. Quanti anni hanno i suoi figli? (Puo' selezionare piu' di un'opzione) Gli intervistati possono
scegliere piu' di un'opzione.
1
0-3 anni
2
4-6 anni
3
7-15 anni
4
16 anni o piu'
D38_03
Valore
Etichetta
(Scelta 3)D38. Quanti anni hanno i suoi figli? (Puo' selezionare piu' di un'opzione) Gli intervistati possono
scegliere piu' di un'opzione.
1
0-3 anni
2
4-6 anni
3
7-15 anni
4
16 anni o piu'
D38_04
Valore
Etichetta
(Scelta 4)D38. Quanti anni hanno i suoi figli? (Puo' selezionare piu' di un'opzione) Gli intervistati possono
scegliere piu' di un'opzione.
1
0-3 anni
2
4-6 anni
3
7-15 anni
4
16 anni o piu'
D39
Valore
Etichetta
D39. Quante auto sono disponibili nella sua famiglia? (Per favore, includa anche le auto aziendali che ha ricevuto
dal suo datore di lavoro e che sono autorizzate per uso personale)
1
Nessuna auto
2
Una auto

353
3
Due auto
4
Tre o piu' auto
D40_01
Valore
Etichetta
D40_1. Approssimativamente, qual e' il suo reddito personale mensile al netto delle tasse?
1
Fino a 500Euro
2
501Euro - 1000Euro
3
1001Euro - 1500Euro
4
1501Euro - 2000Euro
5
2001Euro - 2500Euro
6
2501Euro - 3000Euro
7
3001Euro - 4000Euro
8
4001Euro - 5000Euro
9
5001Euro - 6000Euro
10
6001Euro - 10.000 Euro
11
Piu di 10.001 Euro
D40_02
Valore
Etichetta
D40_2. Approssimativamente, qual e' il reddito mensile della sua famiglia al netto delle tasse? Puo' rispondere a
questa domanda anche se non e' sicuro dell'importo esatto.
1
Fino a 500Euro
2
501Euro - 1000Euro
3
1001Euro - 1500Euro
4
1501Euro - 2000Euro
5
2001Euro - 2500Euro
6
2501Euro - 3000Euro
7
3001Euro - 4000Euro
8
4001Euro - 5000Euro
9
5001Euro - 6000Euro
10
6001Euro - 10.000 Euro
11
Piu di 10.001 Euro
D41
Valore
Etichetta
D41. Come gestisce le sue spese con il suo attuale reddito?
1
Molto bene
2
Abbastanza bene
3
Ne' bene ne' male
4
Abbastanza male
5
Molto male

354
A3.2 The codebook for government members and operators {car,
bike, scooter}-sharing (general codebook) (surveys 4 to 6)
This type of codebook is designed for government members and operators of car sharing, bike-
sharing, and scooter-sharing services. This type of general codebook is presented as follows.
SERVIZIO
Valore
Etichetta
Scelta
1
bike-sharing
2
{car, bike, scooter}-sharing
3
monopattino in condivisione
B1
Valore
Etichetta
Ci sono diverse caratteristiche relative agli spostamenti che potrebbero essere considerate nella scelta del bike-sharing
per effettuare uno spostamento.
B1_01
Valore
Etichetta
B1. Qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è quella MENO IMPORTANTE che potrebbe
influenzare la scelta delle persone? Tempo di percorrenza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_02
Valore
Etichetta
B1. Qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è quella MENO IMPORTANTE che potrebbe
influenzare la scelta delle persone? Distanza da percorrere
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_03
Valore
Etichetta
B1. Qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è quella MENO IMPORTANTE che potrebbe
influenzare la scelta delle persone? Orario di partenza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_04
Valore
Etichetta
B1. Qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è quella MENO IMPORTANTE che potrebbe
influenzare la scelta delle persone? Scopo del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B2_01
Valore

355
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_02
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_03
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_01
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica MENO importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza

356
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_02
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica MENO importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B4
Valore
Etichetta
Ora, esaminiamo l'importanza relativa di alcune caratteristiche del {car, bike, scooter}-sharing. Secondo lei, tra le sei
caratteristiche del {car, bike, scooter}-sharing sopra menzionate che potrebbero influenzare la scelta delle persone,qual è
la caratteristica PIÙ IMPORTANTE e MENO
B4_01
Valore
Etichetta
B4. Costo del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_02
Valore
Etichetta
B4. Comfort del viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_03
Valore
Etichetta
B4. Sicurezza
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto

357
B4_04
Valore
Etichetta
B4. Qualità del servizio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_05
Valore
Etichetta
B4. Sistema rispettoso dell'ambiente
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B4_06
Valore
Etichetta
B4. Facilità di utilizzo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B5_01
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare in che
misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque caratteristiche?
1
1 Ugualeimportanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_02
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare in che
misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque caratteristiche?
1
1 Ugualeimportanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante

358
B5_03
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare in che
misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque caratteristiche?
1
1 Ugualeimportanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_04
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare in che
misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque caratteristiche?
1
1 Ugualeimportanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B5_05
Valore
Etichetta
B5. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare in che
misura considera PIÙ_IMPORTANTE questa caratteristica rispetto alle altre cinque caratteristiche?
1
1 Ugualeimportanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B6_01
Valore

359
Etichetta
B6. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B6_02
Valore
Etichetta
B6. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B6_03
Valore
Etichetta
B6. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B6_04
Valore
Etichetta
B6. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre quattro caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza

360
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B7
Valore
Etichetta
B7. Infine, consideriamo le seguenti due caratteristiche relative al luogo in cui le auto condivise sono effettivamente
disponibili. Secondo lei, qual tra questi due è il fattore PIÙ IMPORTANTE?
1
Disponibilita' del servizio
2
Disponibilita' e accessibilita' del veicolo
B8
Valore
Etichetta
B8. Alla domanda precedente, lei ha scelto ...quale caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE di quella MENO IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B9
Valore
Etichetta
Ora, consideriamo insieme le caratteristiche relative allo spostamento, le caratteristiche del car sharing e la disponibilità e
l'accessibilità che ha valutato separatamente nelle domande precedenti.
B9_01
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale è il MENO
IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento? Caratteristiche
relative al viaggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_02
Valore

361
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale è il MENO
IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_03
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale è il MENO
IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento? Caratteristiche del
{car, bike, scooter}-sharing
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B9_04
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale è il MENO
IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento?
1
Disponibilita' del servizio
2
Disponibilita' e accessibilita' del veicolo
B9_05
Valore
Etichetta
B9. Secondo lei, quale di questi tre gruppi di caratteristiche è complessivamente il PIÙ IMPORTANTE, e quale è il MENO
IMPORTANTE quando si considera di scegliere il {car, bike, scooter}-sharing per uno spostamento? Disponibilità ed
accessibilità
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B10_01
Valore
Etichetta
B10. Alla domanda precedente, lei ha scelto il gruppo ...come gruppo di caratteristiche piu' importanti. Potrebbe per favore
valutare fino a che punto considera il gruppo PIU’_IMPORTANTE più importante degli altri due gruppi?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8

362
9
9 Estremamente piu' importante
B10_02
Valore
Etichetta
B10. Alla domanda precedente, lei ha scelto il gruppo ...come gruppo di caratteristiche piu' importanti. Potrebbe per favore
valutare fino a che punto considera il gruppo PIU’_IMPORTANTE più importante degli altri due gruppi?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B11
Valore
Etichetta
B11. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera
l'altra caratteristica più importante di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
D1
Valore
Etichetta
D1. Secondo la sua opinione, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Tragitti di breve distanza (meno di 5 km)
2
Tragitti di lunga distanza (oltre 5 km)
3
Entrambi
D2
Valore
Etichetta
D2. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Spostamenti di breve durata (meno di 30 minuti)
2
Spostamenti di lunga distanza (oltre 30 min)
3
Entrambi
D3

363
Valore
Etichetta
D3. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Durante le ore di punta
2
Durante le ore non di punta
3
Entrambi
D4_01
Valore
Etichetta
(Scelta 1)D4. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Mattina dei giorni feriali
2
Mattina del fine settimana
3
Sera dei giorni feriali
4
Sera del fine settimana
D4_02
Valore
Etichetta
(Scelta 2)D4. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Mattina dei giorni feriali
2
Mattina del fine settimana
3
Sera dei giorni feriali
4
Sera del fine settimana
D4_03
Valore
Etichetta
(Scelta 3)D4. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare
maggiormente) il {car, bike, scooter}-sharing?
1
Mattina dei giorni feriali
2
Mattina del fine settimana
3
Sera dei giorni feriali
4
Sera del fine settimana
D5
Valore
Etichetta
D5. Secondo lei, quale delle seguenti caratteristiche potrebbe indurre le persone a utilizzare (o utilizzare maggiormente) il
{car, bike, scooter}-sharing?
1
Per viaggi di piacere (ad esempio, far visita ad amici o fare shopping)
2
Per viaggi non di piacere (andare al lavoro/scuola)
3
Entrambi
A3.3. The codebook for users and non-users of shared mobility
services (as a whole) (survey 7)
This type of codebook is designed for users and non-users of shared mobility services (as a
whole). This type of codebook is offered as follows.

364
Genere
Valore
Etichetta
Lei e'
1
Uomo
2
Donna
Comune
Valore
Etichetta
In quale Comune risiedi?
2
Baldissero Torinese
3
Beinasco
4
Borgaro Torinese
5
Cambiano
6
Candiolo
7
Carignano
8
Caselle Torinese
9
Chieri
10
Collegno
11
Druento
12
Grugliasco
13
La Loggia
14
Leini
15
Mappano
16
Moncalieri
17
Nichelino
18
Orbassano
19
Pecetto Torinese
20
Pianezza
21
Pino Torinese
22
Piobesi Torinese
23
Piossasco
24
Rivalta di Torino
25
Rivoli
26
San Mauro Torinese
27
Santena
28
Settimo Torinese
29
Trofarello
30
Venaria Reale
31
Vinovo
32
Volpiano
33
Torino
34
altro
Users_Nonusers
Valore
Etichetta
Tipo
1
users

365
2
Non-users
B1
Valore
Etichetta
B1. Ci sono diverse caratteristiche relative agli spostamenti che potrebbero essere considerate nella scelta della shared-
mobility per effettuare uno spostamento.
B1_01
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Sicurezza delle persone
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_02
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Velocità operativa
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_03
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Accessibilità
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_04
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Facilità d'uso
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_05
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Immagine
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_06
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Comfort
1
Selezioni la caratteristica PIU' importante nella casella qui sotto

366
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_07
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Costo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_08
Valore
Etichetta
B1. Secondo lei, tra le quattro caratteristiche sopra citate, qual è la caratteristica PIÙ IMPORTANTE del viaggio e qual è
quella MENO IMPORTANTE che potrebbe influenzare la sua scelta? Possibilità di trasportare oggetti
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B2_01
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_02
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_03
Valore

367
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_04
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_05
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_06
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2

368
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B2_07
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ...come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre tre caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_01
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_02
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6

369
7
7
8
8
9
9 Estremamentepiu' importante
B3_03
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_04
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
B3_05
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante

370
B3_06
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamentepiu' importante
Intro
Valore
Etichetta
Per favore, risponda alle seguenti domande per determinare la sua opinione sulle caratteristiche che influenzano l'uso del
car-sharing, del bike-sharing e dello scooter-sharing (monopattino in condivisione).
1
Explanation
Q1
Valore
Etichetta
Q1. Quanto si sente sicuro durante i viaggi in car-sharing?
1
1 (Molto insicuro)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto sicuro)
Q2
Valore
Etichetta
Q2. Quanto si sente sicuro nei viaggi in bike-sharing?
1
1 (Molto insicuro)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto sicuro)
Q3
Valore
Etichetta
Q3. Quanto si sente sicuro durante i viaggi in scooter-sharing (monopattino in condivisione)?
1
1 (Molto insicuro)

371
2
2
3
3
4
4
5
5
6
6
7
7 (Molto sicuro)
Q4
Valore
Etichetta
Q4. Come valuterebbe la velocita' di viaggio del servizio di car-sharing?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q5
Valore
Etichetta
Q5. Come valuterebbe la velocita' di viaggio del servizio di bike-sharing?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q6
Valore
Etichetta
Q6. Come valuterebbe la velocita' di viaggio del servizio di scooter-sharing (monopattino in condivisione)?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q7
Valore
Etichetta
Q7. Quanto e' facile o difficile accedere al car-sharing?
1
1 (Molto difficile)
2
2
3
3

372
4
4
5
5
6
6
7
7 (Molto facile)
Q8
Valore
Etichetta
Q8. Quanto e' facile o difficile accedere al bike-sharing?
1
1 (Molto difficile)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto facile)
Q9
Valore
Etichetta
Q9. Quanto e' facile o difficile accedere allo scooter-sharing (monopattino in condivisione)?
1
1 (Molto difficile)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto facile)
Q10
Valore
Etichetta
Q10. Come valuterebbe la facilita' d'uso dei servizi di car-sharing?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q11
Valore
Etichetta
Q11. Come valuterebbe la facilita' d'uso dei servizi di bike sharing?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5

373
6
6
7
7 (Molto buona)
Q12
Valore
Etichetta
Q12. Come valuterebbe la facilita' d'uso dei servizi di scooter-sharing (monopattino in condivisione)?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q13
Valore
Etichetta
Q13. Come valuterebbe il servizio di car-sharing nel complesso?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q14
Valore
Etichetta
Q14. Come valuterebbe il servizio di bike-sharing nel complesso?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)
Q15
Valore
Etichetta
Q15. Come valuterebbe il servizio di scooter-sharing (monopattino in condivisione) nel complesso?
1
1 (Molto scarsa)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto buona)

374
Q16
Valore
Etichetta
Q16. Quanto si sente a suo agio nei viaggi in car-sharing?
1
1 (Molto a disagio)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto a mio agio)
Q17
Valore
Etichetta
Q17. Quanto si sente a suo agio nei viaggi in bike-sharing?
1
1 (Molto a disagio)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto a mio agio)
Q18
Valore
Etichetta
Q18. Quanto si sente a suo agio nei viaggi in scooter-sharing (monopattino in condivisione)?
1
1 (Molto a disagio)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto a mio agio)
Q19
Valore
Etichetta
Q19. Come valuterebbe i costi di utilizzo o di iscrizione ai servizi di car-sharing?
1
1 (Molto costoso)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto economico)
Q20

375
Valore
Etichetta
Q20. Come valuterebbe i costi di utilizzo o di iscrizione ai servizi di bike-sharing?
1
1 (Molto costoso)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto economico)
Q21
Valore
Etichetta
Q21. Come valuterebbe i costi di utilizzo o di iscrizione ai servizi di scooter-sharing (monopattino in condivisione)?
1
1 (Molto costoso)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto economico)
Q22
Valore
Etichetta
Q22. E' difficile o facile trasportare le sue cose quando usa il car-sharing?
1
1 (Molto difficile)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto facile)
Q23
Valore
Etichetta
Q23. E' difficile o facile trasportare le sue cose quando usa il bike-sharing?
1
1 (Molto difficile)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto facile)
Q24
Valore
Etichetta
Q24. E' difficile o facile trasportare le sue cose quando usa lo scooter-sharing (monopattino in condivisione)?

376
1
1 (Molto difficile)
2
2
3
3
4
4
5
5
6
6
7
7 (Molto facile)
A3.4 The codebook for government members about shared
mobility services (as a whole) (survey 8)
This type of codebook is designed for government members and is for shared mobility services
(as a whole). This type of codebook is provided as follows.
B1_01
Valore
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Il numero di viaggi per veicolo al giorno
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_02
Valore
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Gas serra (GHG)
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_03
Valore
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Problemi di parcheggio
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_04
Valore
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Emissione di sostanze inquinanti
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_05
Valore

377
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Integrazione del servizio di mobilita' condivisa con il trasporto
pubblico
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_06
Valore
Etichetta
B1 Secondo lei, tra le sei caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella MENO
IMPORTANTE che potrebbe influenzare la sua scelta?: Tassa sul veicolo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B2_01
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica PIU' importante. Potrebbe per favore valutare fino a
che punto considera questa caratteristica PIÙ IMPORTANTE delle altre cinque caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_02
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica PIU' importante. Potrebbe per favore valutare fino a
che punto considera questa caratteristica PIÙ IMPORTANTE delle altre cinque caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_03
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica PIU' importante. Potrebbe per favore valutare fino a
che punto considera questa caratteristica PIÙ IMPORTANTE delle altre cinque caratteristiche?

378
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_04
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica PIU' importante. Potrebbe per favore valutare fino a
che punto considera questa caratteristica PIÙ IMPORTANTE delle altre cinque caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_05
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica PIU' importante. Potrebbe per favore valutare fino a
che punto considera questa caratteristica PIÙ IMPORTANTE delle altre cinque caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_01
Valore
Etichetta
B3. Inoltre, lei ha scelto ...come caratteristica MENO importante.Potrebbe per favore valutare in che misura considera
le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4

379
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_02
Valore
Etichetta
B3. Inoltre, lei ha scelto ...come caratteristica MENO importante.Potrebbe per favore valutare in che misura considera
le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_03
Valore
Etichetta
B3. Inoltre, lei ha scelto ...come caratteristica MENO importante.Potrebbe per favore valutare in che misura considera
le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_04
Valore
Etichetta
B3. Inoltre, lei ha scelto ...come caratteristica MENO importante.Potrebbe per favore valutare in che misura considera
le altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8

380
9
9 Estremamente piu' importante
A3.5 The codebook for operators of shared mobility services (as a
whole) (survey 9)
This type of codebook is designed for operators of shared mobility services (as a whole). This
type of codebook is presented as follows.
B_pre_01
Valore
Etichetta
(Scelta 1)Che tipo di servizio di mobilita' condivisa viene offerto dalla sua azienda?
1
Car-sharing a flusso libero
2
Car-sharing a prenotazione
3
Bike-sharing a flusso libero
4
Bike-sharing a prenotazione
5
monopattino in condivisione
B_pre_02
Valore
Etichetta
(Scelta 2)Che tipo di servizio di mobilita' condivisa viene offerto dalla sua azienda?
1
Car-sharing a flusso libero
2
Car-sharing a prenotazione
3
Bike-sharing a flusso libero
4
Bike-sharing a prenotazione
5
monopattino in condivisione
B_pre_03
Valore
Etichetta
(Scelta 3)Che tipo di servizio di mobilita' condivisa viene offerto dalla sua azienda?
1
Car-sharing a flusso libero
2
Car-sharing a prenotazione
3
Bike-sharing a flusso libero
4
Bike-sharing a prenotazione
5
monopattino in condivisione
B_pre_04
Valore
Etichetta
(Scelta 4)Che tipo di servizio di mobilita' condivisa viene offerto dalla sua azienda?
1
Car-sharing a flusso libero
2
Car-sharing a prenotazione
3
Bike-sharing a flusso libero
4
Bike-sharing a prenotazione
5
monopattino in condivisione

381
B_pre_05
Valore
Etichetta
(Scelta 5)Che tipo di servizio di mobilita' condivisa viene offerto dalla sua azienda?
1
Car-sharing a flusso libero
2
Car-sharing a prenotazione
3
Bike-sharing a flusso libero
4
Bike-sharing a prenotazione
5
monopattino in condivisione
B1_01
Valore
Etichetta
B1 Secondo lei, tra le cinque caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella
MENO IMPORTANTE che potrebbe influenzare la sua scelta?: Tasso di utilizzo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_02
Valore
Etichetta
B1 Secondo lei, tra le cinque caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella
MENO IMPORTANTE che potrebbe influenzare la sua scelta?: Costi di utilizzo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_03
Valore
Etichetta
B1 Secondo lei, tra le cinque caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella
MENO IMPORTANTE che potrebbe influenzare la sua scelta?: Numero di viaggi per veicolo al giorno
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_04
Valore
Etichetta
B1 Secondo lei, tra le cinque caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella
MENO IMPORTANTE che potrebbe influenzare la sua scelta?: Velocita' operativa
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B1_05
Valore

382
Etichetta
B1 Secondo lei, tra le cinque caratteristiche sopra citate, qual e' la caratteristica PIU' IMPORTANTE e qual e' quella
MENO IMPORTANTE che potrebbe influenzare la sua scelta?: Vita media del veicolo
1
Selezioni la caratteristica PIU' importante nella casella qui sotto
2
Selezioni la caratteristica MENO importante nella casella qui sotto
B2_01
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre quattro caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_02
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre quattro caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_03
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre quattro caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5

383
6
6
7
7
8
8
9
9 Estremamente piu' importante
B2_04
Valore
Etichetta
B2. Alla domanda precedente, lei ha scelto ... come caratteristica piu' importante. Potrebbe per favore valutare fino a che
punto considera questa caratteristica PIÙ IMPORTANTE delle altre quattro caratteristiche?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_01
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
B3_02
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8

384
9
9 Estremamente piu' importante
B3_03
Valore
Etichetta
B3. Inoltre, lei ha scelto ... come caratteristica meno importante. Potrebbe per favore valutare in che misura considera le
altre due caratteristiche più importanti di quella MENO_IMPORTANTE?
1
1 Uguale importanza
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9 Estremamente piu' importante
A3.6 Job positions of government members and operators (surveys
4, 5, 6, 8, and 9)
To better understand the perspectives of government members and operators, it is important to
understand their job position. In this regard, Tables A2 to A9 show the job positions of
government members (surveys 4, 5, 6, and 8) and operators (surveys 4, 5, 6, and 9) according
to the type of shared mobility service.
Table A2: Job position of government members who responded to a survey on shared
mobility services (survey 8).
Shared mobility services (government members)
Respondent ID
The role of government members
IDU_003
Councilor of ecological and digital transition innovation, mobility, and transport
IDU_004
Regional manager for transport and infrastructure investments
IDU_008
Transport and infrastructure planning and programming sector manager
IDU_012
Responsible for European sustainable mobility projects
IDU_013
District president 4 TO
IDU_014
Planning officer
IDU_015
European designer

385
Table A3: Job position of operators of shared mobility services and their type of shared
mobility service (survey 9).
Shared mobility services (Operators)
Respondent
ID
Type of Shared Mobility
The role of operators
IDU_009
Free-floating bike-sharing
Manager of technological services for mobility in one mid/sized city in the
Lombardy region
IDU_010
Scooter-sharing
Managing director of micro-mobility through shared scooters
IDU_011
Scooter-sharing
General manager for Italy and expansion marketing operations for shared scooters
in Stockholm, Milan, Turin, and other cities
IDU_016
Free-floating bike-sharing
Operations manager
IDU_017
Station-based bike-sharing
Project manager
IDU_018
Free-floating car-sharing
Responsible for smart mobility
IDU_019
Station-based car-sharing, Free-
floating car-sharing
Operational office employee
IDU_020
Free-floating car-sharing
Developer of a rental car-sharing business
IDU_021
Scooter-sharing
Regional general manager southern Europe
Table A4: Job position of government members who responded to a survey on car-sharing
services (survey 4).
Car-sharing services (government members)
Respondent ID
The role of government members
IDU_001
Technical manager for European mobility projects
IDU_006
Municipal advisor of the environment commission
IDU_007
Turin council councilor
IDU_008
Transport and infrastructure planning and programming sector manager
Table A5: Job position of operators of car-sharing services and their type of car-sharing
service (survey 4).
Car-sharing services (operators)
Respondent ID
Type of car-sharing
The role of operators
IDU_018
Free-floating car-sharing
Responsible for smart mobility
IDU_019
Station-based car-sharing
Operational office employee
IDU_020
Free-floating car-sharing
Developer of a rental car-sharing business
Table A6: Job position of government members who responded to a survey on bike-sharing
services (survey 5).
Bike-sharing services (government members)
Respondent ID
The role of government members
IDU_002
Director of the transport staff of a metropolitan city in Northern Italy
IDU_005
Officer for mobility, logistics, and citizen services
IDU_012
Responsible for European sustainable mobility projects
IDU_013
District president 4 TO
IDU_014
Planning officer

386
Table A7: Job position of operators of bike-sharing services and their type of bike-sharing
service (survey 5).
Bike-sharing services (operators)
Respondent ID
Type of bike-sharing
The role of operators
IDU_009
Free-floating bike-sharing
Manager of technological services for mobility in one mid/sized city in the Lombardy
region
IDU_016
Free-floating bike-sharing
Operations manager
IDU_017
Station-based bike-
sharing
Project manager
Table A8: Job position of government members who responded to a survey on scooter-
sharing services (survey 6).
Scooter-sharing services (government members)
Respondent ID
The role of government members
IDU_003
Councilor of ecological and digital transition innovation, mobility, and transport
IDU_004
Regional manager for transport and infrastructure investments
IDU_015
European designer
Table A 9: Job position of operators of scooter-sharing services and their type of scooter-
sharing service (survey 6).
Scooter-sharing services (operators)
Respondent
ID
Type of scooter-
sharing
The role of operators
IDU_010
Scooter-sharing
Managing director of micro-mobility through shared scooters
IDU_011
Scooter-sharing
General manager for Italy and expansion marketing operations for shared scooters in
Stockholm, Milan, Turin, and other cities
IDU_021
Scooter-sharing
Regional general manager southern Europe

387
Appendix 4
Appendix 4: Descriptive statistics of the
data set
A4.1 Socio-demographic characteristics of users and non-users of
each of the shared mobility services
The socio-demographic characteristics of survey respondents who are users and non-users of
car-sharing, bike-sharing, and scooter-sharing services are listed in Table A10 (question set C
in surveys 1 to 3).
Table A10: Socio-demographic characteristics of survey respondents (users and non-users
separately) associated with each shared mobility service (question set C in surveys 1 to 3).
Socio-demographic factors
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Gender
Male
37
(48.68%)
64
(50.79%)
49
(65.33%)
60
(47.24%)
44
(57.14%)
51
(40.48%)
Female
39
(51.32%)
62
(49.21%)
26
(34.67%)
67
(52.67%)
33
(42.86%)
75
(59.52%)
Age
18-24
3
(3.95%)
9
(7.14%)
-
4 (3.15%)
3
(3.90%)
4 (3.17%)
25-34
20
(26.32%)
24
(19.05%)
24
(32.00%)
15
(11.81%)
19
(24.68%)
14
(11.11%)
35-44
21
(27.63%)
13
(10.32%)
20
(26.67%)
26
(20.47%)
17
(22.08%)
34
(26.98%)
45-54
21
(27.63%)
26
(20.63%)
10
(13.33%)
37
(29.13%)
4
(5.19%)
43
(34.13%)
55-64
8
(10.53%)
32
(25.40%)
14
(18.67%)
19
(14.96%)
20
(25.97%)
25
(19.84%)
> 64
3
(3.95%)
22
(17.46%)
7
(9.33%)
26
(20.47%)
14
(18.18%)
6 (4.76%)
Education level
Not completed primary
school
-
-
-
-
-
-
Elementary school
-
1
(0.79%)
-
3 (2.36%)
1
(1.30%)
-

388
Socio-demographic factors
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Upper secondary school or
equivalent shorter than three
years
1
(1.32%)
12
(9.52%)
2
(2.67%)
9 (7.09%)
4
(5.19%)
10
(7.94%)
Upper secondary school or
equivalent three years or
more
23
(30.26%)
50
(39.68%)
22
(29.33%)
48
(37.80%)
26
(33.77%)
37
(29.37%)
Post-secondary education,
not college, less than three
years
6
(7.89%)
4
(3.17%)
2
(2.67%)
9 (7.09%)
4
(5.19%)
7 (5.56%)
Post-secondary education,
not college, three years or
more
4
(5.26%)
10
(7.94%)
6
(8.00%)
7 (5.51%)
5
(6.49%)
12
(9.52%)
University less than three
years
3
(3.95%)
4
(3.17%)
-
9 (7.09%)
4
(5.19%)
8 (6.35%)
University 3 years or more
29
(38.16%)
38
(30.16%)
26
(34.67%)
31
(24.41%)
25
(32.47%)
42
(33.33%)
Degree from postgraduate
studies
10
(13.16%)
7
(5.56%)
17
(22.67%)
11
(8.66%)
8
(10.39%)
10
(7.94%)
Marital status
Single
30
(39.47%)
51
(40.48%)
23
(30.67%)
34
(26.77%)
24
(31.17%)
46
(36.51%)
Married or domestic
partnership
46
(60.53%)
75
(59.52%)
52
(69.33%)
93
(73.23%)
53
(68.83%)
80
(63.49%)
Business or
professional status
Entrepreneur/freelancer
7
(9.21%)
5
(3.97%)
14
(18.67%)
11
(8.66%)
6
(7.79%)
9 (7.14%)
Officer/manager
8
(10.53%)
4
(3.17%)
8
(10.67%)
6 (4.72%)
11
(14.29%)
8 (6.35%)
Clerk/trade employee
34
(44.74%)
42
(33.33%)
20
(26.67%)
43
(33.86%)
22
(28.57%)
60
(47.62%)
Worker
4
(5.26%)
8
(6.35%)
7
(9.33%)
8 (6.30%)
2
(2.60%)
10
(7.94%)
Teacher
3
(3.95%)
6
(4.76%)
5
(6.67%)
7 (5.51%)
1
(1.30%)
5 (3.97%)
Representative
1
(1.32%)
2
(1.59%)
2
(2.67%)
1 (0.79%)
1
(1.30%)
1 (0.79%)
Craftsman / trader / operator
3
(3.95%)
2
(1.59%)
3
(4.00%)
2 (1.57%)
6
(7.79%)
2 (1.59%)
Student
4
(5.26%)
11
(8.73%)
2
(2.67%)
4 (3.15%)
7
(9.09%)
6 (4.76%)
Housewife
4
(5.26%)
8
(6.35%)
4
(5.33%)
10
(7.87%)
3
(3.90%)
7 (5.56%)
Retired
2
(2.63%)
24
(19.05%)
3
(4.00%)
19
(14.96%)
16
(20.78%)
5 (3.97%)
Waiting for first job / never
worked
1
(1.32%)
3
(2.38%)
-
2 (1.57%)
-
1 (0.79%)
Unemployed / lost his/her job
2
(2.63%)
6
(4.76%)
5
(6.67%)
13
(10.24%)
-
5 (3.97%)
Other
3
(3.95%)
5
(3.97%)
2
(2.67%)
1 (0.79%)
2
(2.60%)
7 (5.56%)
Number of people,
including
respondents, living in
the home
One person
10
(13.16%)
21
(16.67%)
18
(24.00%)
16
(12.60%)
15
(19.48%)
24
(19.05%)
Two people
23
(30.26%)
41
(32.54%)
23
(30.67%)
51
(40.16%)
30
(38.96%)
47
(37.30%)
Three people
25
(32.89%)
41
(32.54%)
17
(22.67%)
36
(28.35%)
17
(22.08%)
25
(19.84%)
Four people
13
(17.11%)
21
(16.67%)
13
(17.33%)
23
(18.11%)
12
(15.58%)
28
(22.22%)
Five or more people
5
(6.58%)
2
(1.59%)
4
(5.33%)
1 (0.79%)
3
(3.90%)
2 (1.59%)
Number of drivers,
including
respondents, living in
the home
0
-
1
(0.79%)
-
1 (0.79%)
7
(9.09%)
8 (6.35%)
1
21
(27.63%)
38
(30.16%)
36
(48.00%)
36
(28.35%)
22
(28.57%)
35
(27.78%)
2
38
(50.00%)
63
(50.00%)
28
(37.33%)
75
(59.06%)
30
(38.96%)
70
(55.56%)
More than 2
17
(22.37%)
24
(19.05%)
11
(14.67%)
15
(11.81%)
18
(23.38%)
13
(10.32%)
Presence of children
at home
Yes
34
(44.74%)
47
(37.30%)
27
(36.00%)
50
(39.37%)
22
(28.57%)
47
(37.30%)

389
Socio-demographic factors
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
No
42
(55.26%)
79
(62.70%)
48
(64.00%)
77
(60.63%)
55
(71.43%)
79
(62.70%)
The age of the
respondent's
child/children
0-3 years old
917
5*
7*
9*
2*
6*
4-6 years old
6*
6*
7*
15*
4*
6*
7-15 years old
10*
17*
11*
14*
6*
25*
16 years or more
13*
26*
14*
19*
14*
20*
Number of cars
available for use in
respondent's home
No car
7
(9.21%)
6
(4.76%)
8
(10.67%)
10
(7.87%)
7
(9.09%)
14
(11.11%)
One car
31
(40.79%)
61
(48.41%)
41
(54.67%)
61
(48.03%)
38
(49.35%)
58
(46.03%)
Two cars
33
(43.42%)
51
(40.48%)
22
(29.33%)
51
(40.16%)
24
(31.17%)
49
(38.89%)
Three cars or more
5
(6.58%)
8
(6.35%)
4
(5.33%)
5 (3.94%)
8
(10.39%)
5 (3.97%)
Monthly income of
the respondent after
tax
Up to 500 Euros
3
(3.95%)
17
(13.49%)
7
(9.33%)
20
(15.75%)
5
(6.49%)
16
(12.70%)
501 Euros - 1000 Euros
4
(5.26%)
12
(9.52%)
10
(13.33%)
18
(14.17%)
5
(6.49%)
13
(10.32%)
1001 Euros - 1500 Euros
30
(39.47%)
26
(20.63%)
14
(18.67%)
23
(18.11%)
12
(15.58%)
29
(23.02%)
1501 Euros - 2000 Euros
16
(21.05%)
36
(28.57%)
15
(20.00%)
28
(22.05%)
19
(24.68%)
35
(27.78%)
2001 Euros - 2500 Euros
6
(7.89%)
16
(12.70%)
8
(10.67%)
18
(14.17%)
13
(16.88%)
12
(9.52%)
2501 Euros - 3000 Euros
8
(10.53%)
11
(8.73%)
7
(9.33%)
10
(7.87%)
13
(16.88%)
11
(8.73%)
3001 Euros - 4000 Euros
4
(5.26%)
3
(2.38%)
6
(8.00%)
7 (5.51%)
5
(6.49%)
6 (4.76%)
4001 Euros - 5000 Euros
2
(2.63%)
3
(2.38%)
4
(5.33%)
1 (0.79%)
3
(3.90%)
3 (2.38%)
5001 Euros - 6000 Euros
-
-
2
(2.67%)
1 (0.79%)
1
(1.30%)
1 (0.79%)
6001 Euros - 10000 Euros
2
(2.63%)
-
1
(1.33%)
-
1
(1.30%)
-
More than 10,001 Euros
1
(1.32%)
2
(1.59%)
1
(1.33%)
1 (0.79%)
-
-
Respondent's
household monthly
income after tax
Up to 500 Euros
-
4
(3.17%)
4
(5.33%)
7 (5.51%)
1
(1.30%)
7 (5.56%)
501 Euros - 1000 Euros
1
(1.32%)
8
(6.35%)
6
(8.00%)
9 (7.09%)
2
(2.60%)
6 (4.76%)
1001 Euros - 1500 Euros
11
(14.47%)
15
(11.90%)
11
(14.67%)
20
(15.75%)
8
(10.39%)
17
(13.49%)
1501 Euros - 2000 Euros
15
(19.74%)
28
(22.22%)
10
(13.33%)
15
(11.81%)
11
(14.29%)
24
(19.05%)
2001 Euros - 2500 Euros
13
(17.11%)
24
(19.05%)
11
(14.67%)
20
(15.75%)
14
(18.18%)
13
(10.32%)
2501 Euros - 3000 Euros
12
(15.79%)
18
(14.29%)
7
(9.33%)
24
(18.90%)
14
(18.18%)
24
(19.05%)
3001 Euros - 4000 Euros
11
(14.47%)
19
(15.08%)
10
(13.33%)
25
(19.69%)
14
(18.18%)
20
(15.87%)
4001 Euros - 5000 Euros
6
(7.89%)
5
(3.97%)
8
(10.67%)
2 (1.57%)
5
(6.49%)
12
(9.52%)
5001 Euros - 6000 Euros
1
(1.32%)
3
(2.38%)
5
(6.67%)
4 (3.15%)
2
(2.60%)
1 (0.79%)
6001 Euros - 10000 Euros
4
(5.26%)
-
2
(2.67%)
-
5
(6.49%)
2 (1.59%)
17
Respondents could select more than one option, up to three options.

390
Socio-demographic factors
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
More than 10,001 Euros
2
(2.63%)
2
(1.59%)
1
(1.33%)
1 (0.79%)
1
(1.30%)
-
How respondents
manage their
expenses with their
current income
Very good
12
(15.79%)
5
(3.97%)
2
(2.67%)
8 (6.30%)
12
(15.58%)
6 (4.76%)
Fairly good
30
(39.47%)
53
(42.06%)
35
(46.67%)
56
(44.09%)
32
(41.56%)
49
(38.89%)
Neither good nor bad
25
(32.89%)
45
(35.71%)
22
(29.33%)
39
(30.71%)
24
(31.17%)
45
(35.71%)
Pretty bad
9
(11.84%)
15
(11.90%)
13
(17.33%)
15
(11.81%)
7
(9.09%)
22
(17.46%)
Very bad
-
8
(6.35%)
3
(%4.00)
9 (7.09%)
2
(2.60%)
4 (3.17%)
The municipality
where the
respondents live
Grugliasco
2
(2.63%)
3
(2.38%)
-
-
-
2 (1.59%)
Collegno
5
(6.58%)
3
(2.38%)
4
(5.33%)
-
2
(2.60%)
2 (1.59%)
Venaria Reale
2
(2.63%)
1
(0.79%)
2
(2.67%)
1 (0.79%)
2
(2.60%)
2 (1.59%)
Borgaro Torinese
-
1
(0.79%)
-
1 (0.79%)
-
1 (0.79%)
Settimo Torinese
1
(1.32%)
2
(1.59%)
-
2 (1.57%)
-
-
San Mauro Torinese
1
(1.32%)
-
1
(1.33%)
2 (1.57%)
-
2 (1.59%)
Pino Torinese
-
-
-
1 (0.79%)
2
(2.60%)
-
Moncalieri
1
(1.32%)
9
(7.14%)
1
(1.33%)
6 (4.72%)
1
(1.30%)
7 (5.56%)
Pecetto Torinese
-
1
(0.79%)
1
(1.33%)
-
-
-
Nichelino
-
2
(1.59%)
-
2 (1.57%)
1
(1.30%)
2 (1.59%)
Candiolo
-
-
-
-
-
1 (0.79%)
Beinasco
1
(1.32%)
3
(2.38%)
1
(1.33%)
2 (1.57%)
-
2 (1.59%)
Orbassano
-
2
(1.59%)
-
1 (0.79%)
3
(3.90%)
3 (2.38%)
Rivalta di Torino
-
3
(2.38%)
-
1 (0.79%)
1
(1.30%)
2 (1.59%)
Rivoli
3
(3.95%)
-
-
3 (2.36%)
1
(1.30%)
2 (1.59%)
Alpignano
-
-
-
-
-
1 (0.79%)
Pianezza [protetta]
-
3
(2.38%)
-
-
-
1 (0.79%)
Druento
-
-
-
2 (1.57%)
-
1 (0.79%)
Leini
-
2
(1.59%)
-
-
-
1 (0.79%)
Chieri
-
4
(3.17%)
-
1 (0.79%)
2
(2.60%)
4 (3.17%)
Trofarello
-
5
(3.97%)
-
3 (2.36%)
1
(1.30%)
2 (1.59%)
Cambiano
-
2
(1.59%)
-
-
-
-
Santena
-
-
-
1 (0.79%)
-
-
Caselle Torinese
-
1
(0.79%)
-
2 (1.57%)
-
1 (0.79%)
Volpiano
1
(1.32%)
-
-
1 (0.79%)
-
-
Baldissero Torinese
-
-
-
-
-
-
La Loggia
-
-
-
-
-
-
Carignano
-
-
-
-
-
2 (1.59%)
Vinovo
-
-
1
(1.33%)
1 (0.79%)
2
(2.60%)
2 (1.59%)
Piobesi Torinese
-
1
(0.79%)
-
-
-
-
Piossasco
-
3
(2.38%)
-
2 (1.57%)
-
1 (0.79%)

391
Socio-demographic factors
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Torino
59
(77.63%)
75
(59.52%)
64
(85.33%)
92
(72.44%)
59
(76.62%)
82
(65.08%)
Others
-
-
-
-
-
-
A4.2 Routines and daily travel views of users and non-users of each
of the shared mobility services
The routines and daily travel views of survey respondents who are users and non-users of car-
sharing, bike-sharing, and scooter-sharing services are listed in Table A11 (question set B in
surveys 1 to 3).
Table A11: Routines and daily travel views of users and non-users of each shared mobility
service (question set B in surveys 1 to 3).
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Having a driving license
Yes
76
(100.00%)
118
(93.65%)
72 (96.00%)
116
(91.34%)
72
(93.51%)
118
(93.65%)
No
-
8
(6.35%)
3 (4.00%)
11
(8.66%)
5 (6.49%)
8 (6.35%)
Experience using
Currently using
76
(100.00%)
-
75
(100.00%%)
-
77
(100.00%)
-
Used to use it in
the past but not
anymore
-
9
(7.14%)
-
5
(3.94%)
-
18
(14.29%)
Never used it
but being
familiar with it
-
102
(80.95%)
-
106
(83.46%)
-
64
(50.79%)
Not familiar
with its concept
-
15
(11.90%)
-
16
(12.60%)
-
44
(34.92%)
The level of people's
familiarity with the service
(only people who are at least
familiar with it)
1(slightly
Familiar)
2 (2.63%)
32
(28.83%)
-
34
(30.63%)
20
(25.97%)
21
(25.61%)
2
1 (1.32%)
27
(24.32%)
2 (2.67%)
30
(27.03%)
12
(15.58%)
15
(18.29%)
3
21
(27.63%)
34
(30.63%)
20 (26.67%)
27
(24.32%)
18
(23.38%)
30
(36.59%)
4
32
(42.11%)
16
(14.41%)
41 (54.67%)
15
(13.51%)
13
(16.88%)
12
(14.63%)
5 (Very
Familiar)
20
(26.32%)
2
(1.80%)
12 (16.00%)
5
(4.50%)
14
(18.18%)
4 (4.88%)
The name of the service
provider company (answers
only belong to people who are
currently using this service)
Enjoy
3418
-
-
-
-
-
Car2go (Share
Now)
47*
-
-
-
-
-
BlueTorino
2119
-
-
-
-
-
TOBike
-
-
51*
-
-
-
Mobike
-
-
30*
-
-
-
Bird
-
-
-
-
23*
-
BIT mobility
-
-
-
-
39*
-
Dott
-
-
-
-
20*
-
Helbiz An
-
-
-
-
6*
-
Circ
-
-
-
-
-
-
Lime
-
-
-
-
4*
-
Wind
-
-
-
-
11*
-
Link
-
-
-
-
5*
-
Vo i
-
-
-
-
1*
-
18
Respondents could select more than one option, up to three options.
19
Respondents could select more than one option, up to three options.

392
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Others
0 (0.00%)
-
0 (0.00%)
-
0 (0.00%)
-
Pick-up locations near home,
or home being in an
operational area (answers only
belong to people who are at
least familiar with the service)
Yes
50
(65.79%)
43
(38.74%)
55 (73.33%)
47
(42.34%)
44
(57.14%)
41
(50.00%)
No
20
(26.32%)
38
(34.23%)
15 (20.00%)
45
(40.54%)
21
(27.27%)
27
(32.93%)
Not knowing
6 (7.89%)
30
(27.03%)
5 (6.67%)
19
(17.12%)
12
(15.58%)
14
(17.07%)
Pick-up locations near the
most frequent destination of
the trips or destination within
the operational area (answers
only belong to people who are
at least familiar with the
service)
Yes
48
(63.16%)
34
(30.63%)
48 (64.00%)
47
(42.34%)
42
(54.55%)
38
(46.34%)
No
16
(21.05%)
31
(27.93%)
19 (25.33%)
33
(29.73%)
22
(28.57%)
22
(26.83%)
Not knowing
12
(15.79%)
46
(41.44%)
8 (10.67%)
31
(27.93%)
13
(16.88%)
22
(26.83%)
The amount of use of a private
car as a driver (answers only
belong to people who have a
driving license)
Daily
25
(32.89%)
42
(35.59%)
12 (16.67%)
46
(39.66%)
25
(34.72%)
38
(32.20%)
4-6 days a week
13
(17.11%)
17
(14.41%)
14 (19.44%)
17
(14.66%)
12
(16.67%)
20
(16.95%)
1-3 days a week
17
(22.37%)
20
(16.95%)
22 (30.56%)
20
(17.24%)
17
(23.61%)
24
(20.34%)
Once/a few
times a month
9
(11.84%)
12
(10.17%)
6 (8.33%)
10
(8.62%)
7 (9.72%)
7 (5.93%)
Rarely
6 (7.89%)
9
(7.63%)
6 (8.33%)
6
(5.17%)
6 (8.33%)
7 (5.93%)
Never
6 (7.89%)
18
(15.25%)
12 (16.67%)
17
(14.66%)
5 (6.94%)
22
(18.64%)
The amount of use of a private
car as a passenger
Daily
3 (3.95%)
4
(3.17%)
2 (2.67%)
5
(3.94%)
3 (3.90%)
2 (1.59%)
4-6 days a week
8
(10.53%)
8
(6.35%)
5 (6.67%)
14
(11.02%)
7 (9.09%)
6 (4.76%)
1-3 days a week
19
(25.00%)
38
(30.16%)
25 (33.33%)
35
(27.56%)
20
(25.97%)
39
(30.95%)
Once/a few
times a month
21
(27.63%)
26
(20.63%)
14 (18.67%)
23
(18.11%)
17
(22.08%)
21
(16.67%)
Rarely
16
(21.05%)
31
(24.60%)
22 (29.33%)
33
(25.98%)
19
(24.68%)
29
(23.02%)
Never
9
(11.84%)
19
(15.08%)
7 (9.33%)
17
(13.39%)
11
(14.29%)
29
(23.02%)
The amount of use of car-
sharing
Daily
-
-
-
2
(1.57%)
-
-
4-6 days a week
9
(11.84%)
-
6 (8.00%)
-
5 (6.49%)
1 (0.79%)
1-3 days a week
17
(22.37%)
-
12 (16.00%)
5
(3.94%)
14
(18.18%)
6 (4.76%)
Once/a few
times a month
26
(34.21%)
2
(1.59%)
13 (17.33%)
8
(6.30%)
10
(12.99%)
9 (7.14%)
Rarely
24
(31.58%)
7
(5.56%)
15 (20.00%)
27
(21.26%)
15
(19.48%)
25
(19.84%)
Never
-
117
(92.86%)
29 (38.67%)
85
(66.93%)
33
(42.86%)
85
(67.46%)
The amount of use of public
transport
Daily
3 (3.95%)
11
(8.73%)
5 (6.67%)
11
(8.66%)
8
(10.39%)
11
(8.73%)
4-6 days a week
9
(11.84%)
9
(7.14%)
12 (16.00%)
6
(4.72%)
13
(16.88%)
18
(14.29%)
1-3 days a week
14
(18.42%)
15
(11.90%)
20 (26.67%)
12
(9.45%)
16
(20.78%)
12
(9.52%)
Once/a few
times a month
22
(28.95%)
15
(11.90%)
16 (21.33%)
26
(20.47%)
18
(23.38%)
27
(21.43%)
Rarely
19
(25.00%)
40
(31.75%)
19 (25.33%)
43
(33.86%)
17
(22.08%)
37
(29.37%)
Never
9
(11.84%)
36
(28.57%)
3 (4.00%)
29
(22.83%)
5 (6.49%)
21
(16.67%)
The amount of use of
motorcycles/scooters
Daily
3 (3.95%)
-
2 (2.67%)
4
(3.15%)
4 (5.19%)
1 (0.79%)
4-6 days a week
3 (3.95%)
1
(0.79%)
4 (5.33%)
1
(0.79%)
5 (6.49%)
4 (3.17%)
1-3 days a week
4 (5.26%)
4
(3.17%)
8 (10.67%)
4
(3.15%)
13
(16.88%)
4 (3.17%)

393
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Once/a few
times a month
7 (9.21%)
10
(7.94%)
8 (10.67%)
5
(3.94%)
7 (9.09%)
6 (4.76%)
Rarely
11
(14.47%)
12
(9.52%)
8 (10.67%)
12
(9.45%)
8
(10.39%)
17
(13.49%)
Never
48
(63.16%)
99
(78.57%)
45 (60.00%)
101
(79.53%)
40
(51.95%)
94
(74.60%)
The amount of use of a taxi
Daily
-
-
1 (1.33%)
2
(1.57%)
-
-
4-6 days a week
1 (1.32%)
2
(1.59%)
3 (4.00%)
1
(0.79%)
3 (3.90%)
2 (1.59%)
1-3 days a week
4 (5.26%)
4
(3.17%)
6 (8.00%)
3
(2.36%)
11
(14.29%)
5 (3.97%)
Once/a few
times a month
12
(15.79%)
9
(7.14%)
15 (20.00%)
7
(5.51%)
10
(12.99%)
13
(10.32%)
Rarely
23
(30.26%)
40
(31.75%)
24 (32.00%)
49
(38.58%)
28
(36.36%)
43
(34.13%)
Never
36
(47.37%)
71
(56.35%)
26 (34.67%)
65
(51.18%)
25
(32.47%)
63
(50.00%)
The amount of use of a
personal bike
Daily
7 (9.21%)
1
(0.79%)
7 (9.33%)
4
(3.15%)
1 (1.30%)
5 (3.97%)
4-6 days a week
7 (9.21%)
7
(5.56%)
3 (4.00%)
2
(1.57%)
11
(14.29%)
5 (3.97%)
1-3 days a week
15
(19.74%)
12
(9.52%)
17 (22.67%)
11
(8.66%)
11
(14.29%)
7 (5.56%)
Once/a few
times a month
14
(18.42%)
24
(19.05%)
15 (20.00%)
20
(15.75%)
13
(16.88%)
17
(13.49%)
Rarely
10
(13.16%)
29
(23.02%)
14 (18.67%)
30
(23.62%)
13
(16.88%)
34
(26.98%)
Never
23
(30.26%)
53
(42.06%)
19 (25.33%)
60
(47.24%)
28
(36.36%)
58
(46.03%)
The amount of use of bike-
sharing
Daily
1 (1.32%)
1
(0.79%)
5 (6.67%)
-
1 (1.30%)
-
4-6 days a week
3 (3.95%)
2
(1.59%)
8 (10.67%)
2
(1.57%)
2 (2.60%)
3 (2.38%)
1-3 days a week
2 (2.63%)
4
(3.17%)
24 (32.00%)
2
(1.57%)
14
(18.18%)
1 (0.79%)
Once/a few
times a month
15
(19.74%)
9
(7.14%)
16 (21.33%)
3
(2.36%)
7 (9.09%)
4 (3.17%)
Rarely
11
(14.47%)
10
(7.94%)
22 (29.33%)
15
(11.81%)
10
(12.99%)
21
(16.67%)
Never
44
(57.89%)
100
(79.37%)
-
105
(82.68%)
43
(55.84%)
97
(76.98%)
The amount of use of scooter-
sharing
Daily
-
-
2 (2.67%)
1
(0.79%)
2 (2.60%)
-
4-6 days a week
-
7
(5.56%)
5 (6.67%)
-
6 (7.79%)
-
1-3 days a week
8
(10.53%)
2
(1.59%)
11 (14.67%)
1
(0.79%)
11
(14.29%)
-
Once/a few
times a month
11
(14.47%)
8
(6.35%)
6 (8.00%)
6
(4.72%)
31
(40.26%)
-
Rarely
7 (9.21%)
8
(6.35%)
6 (8.00%)
12
(9.45%)
27
(35.06%)
-
Never
50
(65.79%)
101
(80.16%)
45 (60.00%)
107
(84.25%)
-
126
(100.00%)
The amount of use of walking
Daily
38
(50.00%)
48
(38.10%)
43 (57.33%)
56
(44.09%)
37
(48.05%)
62
(49.21%)
4-6 days a week
9
(11.84%)
19
(15.08%)
10 (13.33%)
17
(13.39%)
15
(19.48%)
16
(12.70%)
1-3 days a week
17
(22.37%)
29
(23.02%)
12 (16.00%)
31
(24.41%)
14
(18.18%)
21
(16.67%)
Once/a few
times a month
8
(10.53%)
14
(11.11%)
7 (9.33%)
8
(6.30%)
7 (9.09%)
14
(11.11%)
Rarely
3 (3.95%)
8
(6.35%)
2 (2.67%)
10
(7.87%)
2 (2.60%)
8 (6.35%)
Never
1 (1.32%)
8
(6.35%)
1 (1.33%)
5
(3.94%)
2 (2.60%)
5 (3.97%)
Private car as a
driver
36
(47.37%)
62
(49.21%)
21 (28.00%)
58
(45.67%)
33
(42.86%)
61
(48.41%)

394
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
The mode of transportation
most likely to be used to go to
work or school
Private car as a
passenger
2 (2.63%)
5
(3.97%)
5 (6.67%)
6
(4.72%)
6 (7.79%)
4 (3.17%)
Car-sharing
4 (5.26%)
3
(2.38%)
5 (6.67%)
-
3 (3.90%)
1 (0.79%)
Public transport
12
(15.79%)
26
(20.63%)
11 (14.67%)
27
(21.26%)
18
(23.38%)
27
(21.43%)
Moto/Scooter
4 (5.26%)
2
(1.59%)
4 (5.33%)
4
(3.15%)
4 (5.19%)
3 (2.38%)
Taxi
1 (1.32%)
2
(1.59%)
1 (1.33%)
1
(0.79%)
-
-
Personal bike
9
(11.84%)
3
(2.38%)
7 (9.33%)
4
(3.15%)
1 (1.30%)
5 (3.97%)
Bike-sharing
-
-
1 (1.33%)
-
-
-
Scooter-sharing
-
-
-
-
2 (2.60%)
1 (0.79%)
Walking
6 (7.89%)
15
(11.90%)
19 (25.33%)
14
(11.02%)
5 (6.49%)
18
(14.29%)
Other
2 (2.63%)
8
(6.35%)
1 (1.33%)
13
(10.24%)
5 (6.49%)
6 (4.76%)
The mode of transportation
most likely to be used to visit a
close relative/friends/,
relatives/family
Private car as a
driver
39
(51.32%)
66
(52.38%)
26 (34.67%)
68
(53.54%)
36
(46.75%)
65
(51.59%)
Private car as a
passenger
12
(15.79%)
25
(19.84%)
14 (18.67%)
30
(23.62%)
14
(18.18%)
24
(19.05%)
Car-sharing
1 (1.32%)
3
(2.38%)
4 (5.33%)
-
3 (3.90%)
3 (2.38%)
Public transport
8
(10.53%)
18
(14.29%)
11 (14.67%)
6
(4.72%)
13
(16.88%)
13
(10.32%)
Moto/Scooter
4 (5.26%)
2
(1.59%)
2 (2.67%)
4
(3.15%)
3 (3.90%)
1 (0.79%)
Taxi
-
-
1 (1.33%)
1
(0.79%)
1 (1.30%)
2 (1.59%)
Personal bike
5 (6.58%)
3
(2.38%)
4 (5.33%)
2
(1.57%)
2 (2.60%)
3 (2.38%)
Bike-sharing
-
-
3 (4.00%)
1
(0.79%)
-
1 (0.79%)
Scooter-sharing
-
-
-
-
1 (1.30%)
-
Walking
7 (9.21%)
8
(6.35%)
9 (12.00%)
14
(11.02%)
4 (5.19%)
11
(8.73%)
Other
-
1
(0.79%)
1 (1.33%)
1
(0.79%)
-
3 (2.38%)
The mode of transport most
likely to be used to run an
errand in the city center
Private car as a
driver
14
(18.42%)
50
(39.68%)
10 (13.33%)
39
(30.71%)
15
(19.48%)
34
(26.98%)
Private car as a
passenger
1 (1.32%)
8
(6.35%)
6 (8.00%)
14
(11.02%)
4 (5.19%)
5 (3.97%)
Car-sharing
12
(15.79%)
1
(0.79%)
8 (10.67%)
2
(1.57%)
9
(11.69%)
3 (2.38%)
Public transport
26
(34.21%)
39
(30.95%)
16 (21.33%)
40
(31.50%)
28
(36.36%)
52
(41.27%)
Moto/Scooter
3 (3.95%)
5
(3.97%)
2 (2.67%)
5
(3.94%)
2 (2.60%)
5 (3.97%)
Taxi
2 (2.63%)
2
(1.59%)
-
3
(2.36%)
-
-
Personal bike
6 (7.89%)
4
(3.17%)
5 (6.67%)
3
(2.36%)
7 (9.09%)
2 (1.59%)
Bike-sharing
1 (1.32%)
2
(1.59%)
6 (8.00%)
1
(0.79%)
-
-
Scooter-sharing
2 (2.63%)
2
(1.59%)
2 (2.67%)
-
4 (5.19%)
-
Walking
9
(11.84%)
12
(9.52%)
20 (26.67%)
17
(13.39%)
8
(10.39%)
23
(18.25%)
Other
-
1
(0.79%)
-
3
(2.36%)
-
2 (1.59%)
The mode of transport most
likely to be used to go out for
dinner
Private car as a
driver
31
(40.79%)
58
(46.03%)
26 (34.67%)
60
(47.24%)
37
(48.05%)
58
(46.03%)
Private car as a
passenger
14
(18.42%)
32
(25.40%)
12 (16.00%)
40
(31.50%)
14
(18.18%)
30
(23.81%)
Car-sharing
8
(10.53%)
3
(2.38%)
8 (10.67%)
-
3 (3.90%)
6 (4.76%)

395
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Public transport
9
(11.84%)
15
(11.90%)
7 (9.33%)
7
(5.51%)
9
(11.69%)
8 (6.35%)
Moto/Scooter
-
3
(2.38%)
2 (2.67%)
6
(4.72%)
3 (3.90%)
5 (3.97%)
Taxi
4 (5.26%)
4
(3.17%)
2 (2.67%)
1
(0.79%)
3 (3.90%)
3 (2.38%)
Personal bike
2 (2.63%)
4
(3.17%)
2 (2.67%)
2
(1.57%)
-
-
Bike-sharing
1 (1.32%)
-
4 (5.33%)
-
1 (1.30%)
-
Scooter-sharing
2 (2.63%)
-
-
-
2 (2.60%)
-
Walking
45
(6.58%)
3
(2..38%)
12 (16.00%)
7
(5.51%)
5 (6.49%)
12
(9.25%)
Other
-
4
(3.17%)
-
4
(3.15%)
-
4 (3.17%)
The mode of transport most
likely to be used to take an
excursion in nice weather
Private car as a
driver
33
(43.42%)
57
(45.24%)
24 (32.00%)
53
(41.73%)
28
(36.36%)
48
(38.10%)
Private car as a
passenger
14
(18.42%)
30
(23.81%)
13 (17.33%)
33
(25.98%)
16
(20.78%)
29
(23.02%)
Car-sharing
4 (5.26%)
2
(1.59%)
3 (4.00%)
1
(0.79%)
3 (3.90%)
2 (1.59%)
Public transport
10
(13.16%)
8
(6.35%)
5 (6.67%)
5
(3.94%)
4 (5.19%)
6 (4.76%)
Moto/Scooter
6 (7.89%)
2
(1.59%)
5 (6.67%)
3
(2.36%)
3 (3.90%)
7 (5.56%)
Taxi
-
3
(2.38%)
1 (1.33%)
1
(0.79%)
-
-
Personal bike
3 (3.95%)
7
(5.56%)
9 (12.00%)
8
(6.30%)
6 (7.79%)
9 (7.14%)
Bike-sharing
1 (1.32%)
1
(0.79%)
4 (5.33%)
2
(1.57%)
3 (3.90%)
-
Scooter-sharing
-
-
-
-
3 (3.90%)
1 (0.79%)
Walking
5 (6.58%)
15
(11.90%)
11 (14.67%)
19
(14.96%)
11
(14.29%)
21
(16.67%)
Other
-
1
(0.79%)
-
2
(1.57%)
-
3 (2.38%)
The mode of transport most
likely to be used to visit a
shopping center
Private car as a
driver
41
(53.95%)
71
(56.35%)
32 (42.67%)
67
(52.76%)
40
(51.95%)
74
(58.73%)
Private car as a
passenger
6 (7.89%)
20
(15.87%)
9 (12.00%)
29
(22.83%)
11
(14.29%)
13
(10.32%)
Car-sharing
7 (9.21%)
2
(1.59%)
5 (6.67%)
-
3 (3.90%)
3 (2.38%)
Public transport
13
(17.11%)
19
(15.08%)
10 (13.33%)
12
(9.45%)
10
(12.99%)
21
(16.67%)
Moto/Scooter
-
1
(0.79%)
5 (6.67%)
3
(2.36%)
2 (2.60%)
4 (3.17%)
Taxi
3 (3.95%)
4
(3.17%)
4 (5.33%)
1
(0.79%)
-
-
Personal bike
3 (3.95%)
4
(3.17%)
1 (1.33%)
3
(2.36%)
4 (5.19%)
1 (0.79%)
Bike-sharing
-
-
4 (5.33%)
1
(0.79%)
2 (2.60%)
1 (0.79%)
Scooter-sharing
1 (1.32%)
1
(0.79%)
-
-
-
-
Walking
1 (1.32%)
2
(1.59%)
2 (2.67%)
8
(6.30%)
5 (6.49%)
5 (3.97%)
Other
1 (1.32%)
2
(1.59%)
3 (4.00%)
3
(2.36%)
-
4 (3.17%)
The mode of transport most
likely to be used to go to
smaller shops
Private car as a
driver
18
(23.68%)
35
(27.78%)
11 (14.67%)
34
(26.77%)
18
(23.38%)
32
(25.40%)
Private car as a
passenger
-
12
(9.52%)
2 (2.67%)
12
(9.45%)
6 (7.79%)
5 (3.97%)
Car-sharing
6 (7.89%)
-
3 (4.00%)
3
(2.36%)
4 (5.19%)
3 (2.38%)
Public transport
10
(13.16%)
15
(11.90%)
12 (16.00%)
4
(3.15%)
13
(16.88%)
23
(18.25%)
Moto/Scooter
3 (3.95%)
2
(1.59%)
5 (6.67%)
6
(4.72%)
3 (3.90%)
3 (2.38%)

396
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Taxi
2 (2.63%)
2
(1.59%)
-
1
(0.79%)
1 (1.30%)
-
Personal bike
6 (7.89%)
11
(8.73%)
4 (5.33%)
3
(2.36%)
4 (5.19%)
2 (1.59%)
Bike-sharing
-
2
(1.59%)
6 (8.00%)
1
(0.79%)
-
1 (0.79%)
Scooter-sharing
3 (3.95%)
1
(0.79%)
2 (2.67%)
-
2 (2.60%)
-
Walking
27
(35.53%)
46
(36.51%)
30 (40.00%)
62
(48.82%)
26
(33.77%)
55
(43.65%)
Other
1 (1.32%)
-
-
1
(0.79%)
-
2 (1.59%)
The mode of transport most
likely to be used for weekend
activities
Private car as a
driver
33
(43.42%)
57
(45.24%)
23 (30.67%)
51
(40.16%)
31
(40.26%)
56
(44.44%)
Private car as a
passenger
12
(15.79%)
23
(18.25%)
9 (12.00%)
35
(27.56%)
8
(10.39%)
23
(18.25%)
Car-sharing
3 (3.95%)
3
(2.38%)
6 (8.00%)
1
(0.79%)
6 (7.79%)
2 (1.59%)
Public transport
7 (9.21%)
10
(7.94%)
8 (10.67%)
2
(1.57%)
10
(12.99%)
15
(11.90%)
Moto/Scooter
1 (1.32%)
-
3 (4.00%)
5
(3.94%)
3 (3.90%)
5 (3.97%)
Taxi
1 (1.32%)
2
(1.59%)
2 (2.67%)
1
(0.79%)
-
-
Personal bike
6 (7.89%)
8
(6.35%)
2 (2.67%)
7
(5.51%)
2 (2.60%)
6 (4.76%)
Bike-sharing
4 (5.26%)
-
8 (10.67%)
-
-
-
Scooter-sharing
2 (2.63%)
2
(1.59%)
1 (1.33%)
-
5 (6.49%)
-
Walking
5 (6.58%)
18
(14.29%)
13 (17.33%)
22
(17.32%)
12
(15.58%)
15
(11.90%)
Other
2 (2.63%)
3
(2.38%)
-
3
(2.36%)
-
4 (3.17%)
The incentive to use car-
sharing (or more use)
Availability
near my
home/work
37*
43*
-
-
-
-
Reduction in
costs
38*
54*
-
-
-
-
More
sustainable
travel
23*
32*
-
-
-
-
Increased
comfort during
travel
12*
16*
-
-
-
-
The convenience
of having it only
when needed
22*
49*
-
-
-
-
Avoiding
responsibilities
related to
maintenance
and repairs
16*
50*
-
-
-
-
The incentive to use bike-
sharing/scooter-sharing (or
more use)
Availability
near my
home/work
-
-
40*
49*
36*
44*
Reduction in
costs
-
-
25*
31*
25*
41*
More
sustainable
travel
-
-
26*
45*
21*
32*
Increased
comfort during
travel
-
-
16*
16*
13*
14*
The convenience
of having it only
when needed
-
-
26*
52*
28*
41*
Avoiding
responsibilities
related to
-
-
14*
28*
18*
38*

397
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
maintenance
and repairs
Smooth track
without slope
-
-
1*
2*
9*
4*
Weather conditions that lead
to car-sharing/bike-
sharing/scooter-sharing use
(answers only belong to people
who are at least familiar with
the service)
Bad weather
(e.g., rainy or
snowy weather)
4320
48*
2*
3*
3*
2*
Good weather
(e.g., sunny
weather)
8*
5*
44*
75*
47*
53*
Scorching
weather
6*
6*
7*
3*
12*
4*
Favorable air
temperature
6*
4*
30*
40*
26*
27*
Freezing
weather
19*
27*
6*
9*
5*
2*
High humidity
level
7*
5*
2*
1*
2*
3*
Favorable
humidity level
4321
2*
2*
5*
5*
2*
High air
pollution
5*
37*
7*
5*
5*
5*
Low air
pollution
4*
5*
20*
21*
11*
14*
In winter
27*
33*
1*
2*
2*
1*
In spring
4*
11*
32*
56*
26*
32*
In summer
7*
8*
27*
30*
24*
36*
In autumn
6*
12*
1*
9*
3*
2*
Travel distance that may cause
the use of the service
Travel less than
5 km
25
(32.89%)
32
(25.40%)
42 (56.00%)
89
(70.08%)
53
(68.83%)
90
(71.43%)
Travel 5 km or
more
28
(36.84%)
44
(34.92%)
12 (16.00%)
9
(7.09%)
9
(11.69%)
14
(11.11%)
Both
23
(30.26%)
50
(39.68%)
21 (28.00%)
29
(22.83%)
15
(19.48%)
22
(17.46%)
Travel time that may cause the
use of the service
Travel less than
30 min
43
(56.58%)
46
(36.51%)
50 (66.67%)
92
(72.44%)
59
(76.62%)
100
(79.37%)
Travel 30 min or
more
16
(21.05%)
35
(27.78%)
10 (13.33%)
16
(12.60%)
6 (7.79%)
7 (5.56%)
Both
17
(22.37%)
45
(35.71%)
15 (20.00%)
19
(14.96%)
12
(15.58%)
19
(15.08%)
Departure time (hour) that
may cause the use of the service
Travel during
peak hours
16
(21.05%)
43
(34.13%)
29 (38.67%)
48
(37.80%)
36
(46.75%)
50
(39.68%)
Travel during
off-peak hours
33
(43.42%)
36
(28.57%)
21 (28.00%)
43
(33.86%)
24
(31.17%)
28
(22.22%)
Both
27
(35.53%)
47
(37.30%)
25 (33.33%)
36
(28.35%)
17
(22.08%)
48
(38.10%)
Departure time (day) that may
cause the use of the service
Travel on a
weekday
morning
30*
61*
43*
68*
44*
80*
Travel on a
weekend
morning
25*
32*
31*
71*
39*
47*
Travel on a
weekday
evening
30*
46*
17*
25*
15*
28*
Travel on a
weekend
evening
30*
36*
12*
15*
12*
18*
The trip purpose that may
cause the use of the service
Travel for
leisure (e.g.,
vising friends or
shopping)
19
(25.00%)
36
(28.57%)
24 (32.00%)
55
(43.31%)
31
(40.26%)
31
(24.60%)
Travel for non-
leisure (going to
work/school)
27
(35.53%)
52
(41.27%)
23 (30.67%)
36
(28.35%)
20
(25.97%)
55
(43.65%)
20
Respondents could select more than one option, up to three options.
21
Respondents could select more than one option, up to three options.

398
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
Both
30
(39.47%)
38
(30.16%)
28 (37.33%)
36
(28.35%)
26
(33.77%)
40
(31.75%)
It is possible for me to use car-
sharing/bike-sharing/scooter-
sharing for my regular trips
(According to perceptions of
respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
10
(13.16%)
3
(33.33%)
6 (8.00%)
1
(20.00%)
3 (3.90%)
-
2
13
(17.11%)
-
16 (21.33%)
1
(20.00%)
10
(12.99%)
-
3
9
(11.84%)
1
(11.11%)
5 (6.67%)
-
16
(20.78%)
3
(16.67%)
4
15
(19.74%)
1
(11.11%)
15 (20.00%)
1
(20.00%)
19
(24.68%)
4
(22.22%)
5
18
(23.68%)
1
(11.11%)
18 (24.00%)
2
(40.00%)
17
(22.08%)
5
(27.78%)
6
5 (6.58%)
2
(22.22%)
8 (10.67%)
-
9
(11.69%)
4
(22.22%)
7 (Strongly
agree)
6 (7.89%)
1
(11.11%)
7 (9.33%)
-
3 (3.90%)
2
(11.11%)
I am sure I can choose car-
sharing/bike-sharing/scooter-
sharing for my regular trips
during the next week
(According to perceptions of
respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
20
(26.32%)
3
(33.33%)
18 (24.00%)
2
(40.00%)
10
(12.99%)
1 (5.56%)
2
6 (7.89%)
-
16 (21.33%)
2
(40.00%)
7 (9.09%)
-
3
15
(19.74%)
2
(22.22%)
7 (9.33%)
-
6 (7.79%)
3
(16.67%)
4
9
(11.84%)
1
(11.11%)
11 (14.67%)
-
15
(19.48%)
4
(22.22%)
5
16
(21.05%)
3
(33.33%)
12 (16.00%)
-
19
(24.68%)
4
(22.22%)
6
7 (9.21%)
-
5 (6.67%)
-
20
(25.97%)
5
(27.78%)
7 (Strongly
agree)
3 (3.95%)
-
6 (8.00%)
1
(20.00%)
-
1 (5.56%)
The car-sharing/bike-
sharing/scooter-sharing
service is a useful mode of
transport (According to
perceptions of respondents)
(answers only belong to people
who have experience with the
service)
1 (Strongly
disagree)
-
1
(11.11%)
1 (1.33%)
-
2 (2.60%)
-
2
1 (1.32%)
-
6 (8.00%)
-
1 (1.30%)
-
3
7 (9.21%)
-
4 (5.33%)
1
(20.00%)
4 (5.19%)
1 (5.56%)
4
14
(18.42%)
4
(44.44%)
16 (21.33%)
-
16
(20.78%)
4
(22.22%)
5
12
(15.79%)
-
24 (32.00%)
-
22
(28.57%)
4
(22.22%)
6
20
(26.32%)
2
(22.22%)
11 (14.67%)
1
(20.00%)
25
(32.47%)
5
(27.78%)
7 (Strongly
agree)
22
(28.95%)
2
(22.22%)
13 (17.33%)
3
(60.00%)
7 (9.09%)
4
(22.22%)
Car-sharing/bike-
sharing/scooter-sharing helps
me to accomplish activities that
are important to me
(According to perceptions of
respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
4 (5.26%)
3
(33.33%)
7 (9.33%)
1
(20.00%)
4 (5.19%)
7
(38.89%)
2
11
(14.47%)
-
11 (14.67%)
1
(20.00%)
14
(18,18%)
8
(44.44%)
3
9
(11.84%)
1
(11.11%)
11 (14.67%)
1
(20.00%)
19
(24.68%)
3
(16.67%)
4
16
(21.05%)
3
(33.33%)
18 (24.00%)
-
18
(23.38%)
-
5
19
(25.00%)
2
(22.22%)
15 (20.00%)
1
(20.00%)
16
(20.78%)
-
6
11
(14.47%)
-
9 (12.00%)
1
(20.00%)
5 (6.49%)
-
7 (Strongly
agree)
6 (7.89%)
-
4 (5.33%)
-
1 (1.30%)
-
Learning how to use car-
sharing/bike-sharing/scooter-
sharing was easy for me
(According to perceptions of
respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
1 (1.32%)
1
(11.11%)
3 (4.00%)
1
(20.00%)
4 (5.19%)
2
(11.11%)
2
5 (6.58%)
-
3 (4.00%)
-
4 (5.19%)
-
3
6 (7.89%)
-
8 (10.67%)
-
17
(22.08%)
2
(11.11%)
4
11
(14.47%)
1
(11.11%)
18 (24.00%)
1
(20.00%)
17
(22.08%)
2
(11.11%)
5
20
(26.32%)
2
(22.22%)
15 (20.00%)
1
(20.00%)
15
(19.48%)
4
(22.22%)
6
14
(18.42%)
4
(44.44%)
14 (18.67%)
1
(20.00%)
12
(15.58%)
5
(27.78%)

399
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
7 (Strongly
agree)
19
(25.00%)
1
(11.11%)
14 (18.67%)
1
(20.00%)
8
(10.39%)
3
(16.67%)
I find car-sharing/bike-
sharing/scooter-sharing easy to
use (According to perceptions
of respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
2 (2.63%)
-
5 (6.67%)
-
-
1 (5.56%)
2
1 (1.32%)
-
2 (2.67%)
-
2 (2.60%)
2
(11.11%)
3
9
(11.84%)
-
5 (6.675)
-
6 (7.79%)
7
(38.89%)
4
14
(18.42%)
4
(44.44%)
17 (22.67%)
2
(40.00%)
18
(23.38%)
4
(22.22%)
5
16
(21.05%)
1
(11.11%)
15 (20.00%)
2
(40.00%)
21
(27.27%)
2
(11.11%)
6
18
(23.68%)
3
(33.33%)
20 (26.67%)
-
22
(28.57%)
1 (5.56%)
7 (Strongly
agree)
16
(21.05%)
1
(11.11%)
11 (14.67%)
1
(20.00%)
8
(10.39%)
1 (5.56%)
It is difficult to book a
car/bike/scooter at the car-
sharing/bike-sharing/scooter-
sharing website/app
(According to perceptions of
respondents) (answers only
belong to people who have
experience with the service)
1 (Strongly
disagree)
17
(22.37%)
3
(33.33%)
9 (12.00%)
2
(40.00%)
8
(10.39%)
-
2
16
(21.05%)
1
(11.11%)
14 (18.67%)
1
(20.00%)
14
(18.18%)
4
(22.22%)
3
10
(13.16%)
-
7 (9.33%)
1
(20.00%)
15
(19.48%)
7
(38.89%)
4
10
(13.16%)
2
(22.22%)
20 (26.67%)
-
12
(15.58%)
6
(33.33%)
5
11
(14.47%)
1
(11.11%)
14 (18.67%)
-
17
(22.08%)
1 (5.56%)
6
8
(10.53%)
2
(22.22%)
7 (9.33%)
-
9
(11.69%)
-
7 (Strongly
agree)
4 (5.26%)
-
4 (5.33%)
1
(20.00%)
2 (2.60%)
-
People who are important to
me think that I should use it
more often instead of other
modes of transportation
(answers only belong to people
who have experience with the
service)
1 (Strongly
disagree)
18
(23.68%)
4
(44.44%)
17 (22.67%)
2
(40.00%)
14
(18.18%)
3
(16.67%)
2
11
(14.47%)
1
(11.11%)
10 (13.33%)
-
4 (5.19%)
1 (5.56%)
3
11
(14.47%)
1
(11.11%)
6 (8.00%)
1
(20.00%)
19
(24.68%)
2
(11.11%)
4
17
(22.37%)
2
(22.22%)
23 (30.67%)
-
8
(10.39%)
4
(22.22%)
5
7 (9.21%)
-
12 (16.00%)
1
(20.00%)
10
(12.99%)
3
(16.67%)
6
7 (9.21%)
1
(11.11%)
5 (6.67%)
1
(20.00%)
17
(22.08%)
4
(22.22%)
7 (Strongly
agree)
5 (6.58%)
-
2 (2.67%)
-
5 (6.49%)
1 (5.56%)
People who are important to
me like that I use it (answers
only belong to people who have
experience with the service)
1 (Strongly
disagree)
13
(17.11%)
3
(33.33%)
9 (12.00%)
1
(20.00%)
11
(14.29%)
2
(11.11%)
2
6 (7.89%)
-
6 (8.00%)
-
8
(10.39%)
-
3
12
(15.79%)
2
(22.22%)
7 (9.33%)
-
9
(11.69%)
2
(11.11%)
4
16
(21.05%)
1
(11.11%)
15 (20.00%)
1
(20.00%)
22
(28.57%)
3
(16.67%)
5
19
(25.00%)
1
(11.11%)
21 (28.00%)
1
(20.00%)
17
(22.08%)
4
(22.22%)
6
7 (9.21%)
2
(22.22%)
10 (13.33%)
2
(40.00%)
9
(11.69%)
6
(33.33%)
7 (Strongly
agree)
3 (3.95%)
-
7 (9.33%)
-
1 (1.30%)
1 (5.56%)
People who are important to
me agree with my use of it
(answers only belong to people
who have experience with the
service)
1 (Strongly
disagree)
11
(14.47%)
2
(22.22%)
6 (8.00%)
-
7 (9.09%)
-
2
3 (3.95%)
-
7 (9.33%)
1
(20.00%)
1 (1.30%)
1 (5.56%)
3
11
(14.47%)
-
5 (6.67%)
0
(20.00%)
4 (5.19%)
2
(11.11%)
4
17
(22.37%)
3
(33.33%)
19 (25.33%)
3
(60.00%)
20
(25.97%)
5
(27.78%)
5
16
(21.05%)
3
(33.33%)
19 (25.33%)
1
(20.00%)
24
(31.17%)
3
(16.67%)

400
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
6
13
(17.11%)
-
9 (12.00%)
-
15
(19.48%)
5
(27.78%)
7 (Strongly
agree)
5 (6.58%)
1
(11.11%)
10 (13.33%)
-
6 (7.79%)
2
(11.11%)
People who are important to
me think that I should use it
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
34
(33.33%)
-
46
(43.40%)
-
27
(42.19%)
2
-
18
(17.65%)
-
13
(12.26%)
-
6 (9.38%)
3
-
7
(6.86%)
-
12
(11.32%)
-
10
(15.63%)
4
-
29
(28.43%)
-
20
(18.87%)
-
12
(18.75%)
5
-
9
(8.82%)
-
13
(12.26%)
-
4 (6.25%)
6
-
3
(2.94%)
-
-
-
5 (7.81%)
7 (Strongly
agree)
-
2
(1.96%)
-
2
(1.89%)
-
-
People who are important to
me would like me to use it
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
37
(36.27%)
-
51
(48.11%)
-
25
(39.06%)
2
-
16
(15.69%)
-
12
(11.32%)
-
8
(12.50%)
3
-
10
(9.80%)
-
8
(7.55%)
-
9
(14.06%)
4
-
23
(22.25%)
-
19
(17.92%)
-
14
(21.88%)
5
-
9
(8.82%)
-
11
(10.38%)
-
2 (3.13%)
6
-
5
(4.90%)
-
4
(3.77%)
-
4 (6.25%)
7 (Strongly
agree)
-
2
(1.96%)
-
1
(0.94%)
-
2 (3.13%)
People who are important to
me would agree if I used it
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
20
(19.61%)
-
28
(26.42%)
-
15
(23.44%)
2
-
11
(10.78%)
-
6
(5.66%)
-
6 (9.38%)
3
-
9
(8.82%)
-
7
(6.60%)
-
10
(15.63%)
4
-
24
(23.53%)
-
30
(28.30%)
-
15
(23.44%)
5
-
22
(21.57%)
-
17
(16.04%)
-
6 (9.38%)
6
-
8
(7.84%)
-
8
(7.55%)
-
7
(10.94%)
7 (Strongly
agree)
-
8
(7.84%)
-
10
(9.43%)
-
5 (7.81%)
My support for the
implementation of it in society
(answers only belong to people
who are at least familiar with
the service)
1 (Very low)
3 (3.95%)
36
(32.43%)
6 (8.00%)
24
(21.62%)
15
(19.48%)
17
(20.73%)
2
9
(11.84%)
14
(12.61%)
1 (13.33%)
22
(19.82%)
5 (6.49%)
13
(15.85%)
3
9
(11.84%)
6
(5.41%)
6 (8.00%)
14
(12.61%)
9
(11.69%)
12
(14.63%)
4
17
(22.37%)
20
(18.02%)
12 (16.00%)
12
(10.81%)
17
(22..08%)
13
(15.85%)
5
20
(26.32%)
20
(18.02%)
16 (21.33%)
19
(17.12%)
11
(14.29%)
15
(18.29%)
6
9
(11.84%)
11
(9.91%)
12 (16.00%)
11
(9.91%)
14
(18.18%)
11
(13.41%)
7 (Very high)
9
(11.84%)
4
(3.60%)
13 (17.33%)
9
(8.11%)
6 (7.79%)
1 (1.22%)
Overall, my view of it (answers
only belong to people who are
at least familiar with the
service)
1 (Very
negative)
1 (1.32%)
12
(10.81%)
3 (4.00%)
2
(1.80%)
12
(15.58%)
8 (9.76%)
2
4 (5.26%)
4
(3.60%)
5 (6.67%)
5
(4.50%)
2 (2.60%)
9
(10.98%)
3
4 (5.26%)
7
(6.31%)
1 (1.33%)
17
(15.32%)
10
(12.99%)
15
(18.29%)
4
14
(18.42%)
32
(28.83%)
12 (16.00%)
20
(18.02%)
13
(16.88%)
17
(20.73%)

401
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
5
16
(21.05%)
28
(25.23%)
18 (24.00%)
30
(27.03%)
20
(25.97%)
17
(20.73%)
6
19 (25%)
17
(15.32%)
19 (25.33%)
14
(12.61%)
10
(12.99%)
10
(12.20%)
7 (Very positive)
18
(23.68%)
11
(9.91%)
17 (22.67%)
23
(20.72%)
10
(12.99%)
6 (7.32%)
Using it is relatively enjoyable
(answers only belong to people
who have experience with the
service)
1 (Strongly
disagree)
1 (1.64%)
-
2 (3.33%)
-
3 (4.84%)
-
2
3 (4.92%)
-
4 (6.67%)
1
(20.00%)
1 (1.61%)
-
3
5 (8.20%)
1
(14.29%)
5 (8.33%)
-
5 (8.06%)
-
4
17
(27.87%)
2
(28.57%)
13 (21.67%)
-
12
(19.35%)
-
5
10
(16.39%)
2
(28.57%)
15 (25.00%)
2
(40.00%)
20
(32.26%)
4
(23.53%)
6
13
(21.31%)
2
(28.57%)
14 (23.33%)
2
(40.00%)
14
(22.58%)
7
(41.18%)
7 (Strongly
agree)
12
(19.67%)
0 (0.00)
7 (11.67%)
-
7
(11.29%)
6
(35.29%)
Using it is relatively
environmentally friendly
(answers only belong to people
who have experience with the
service)
1 (Strongly
disagree)
-
0.0-
-
0
(0..00%)
2 (3.23%)
-
2
1 (1.64%)
0.0-
4 (6.67%)
-
-
-
3
4 (6.56%)
0.0-
4 (6.67%)
1
(20.00%)
7
(11.29%)
-
4
14
(22.95%)
2
(28.57%)
11 (18.33%)
-
6 (9.68%)
-
5
14
(22.95%)
3
(42.86%)
11 (18.33%)
-
18
(29.03%)
5
(29.41%)
6
20
(32.79%)
-
13 (21.67%)
1
(20.00%)
21
(33.87%)
7
(41.18%)
7 (Strongly
agree)
8
(13.11%)
2
(28.57%)
17 (28.33%)
3
(60.00%)
8
(12.90%)
5
(29.41%)
The impact of health concerns
due to the Covid-19 pandemic
has reduced my use (answers
only belong to people who have
experience with the service)
1 (Strongly
disagree)
16
(26.23%)
3
(42.86%)
15 (25.00%)
1
(20.00%)
9
(14.52%)
-
2
4 (6.56%)
1
(14.29%)
6 (10.00%)
2
(40.00%)
12
(19.35%)
1 (5.88%)
3
4 (6.56%)
-
3 (5.00%)
1
(20.00%)
4 (6.45%)
5
(29.41%)
4
13
(21.31%)
1
(14.29%)
15 (25.00%)
-
11
(17.74%)
7
(41.18%)
5
9
(14.75%)
-
12 (20.00%)
-
11
(17.74%)
4
(23.53%)
6
4 (6.56%)
1
(14.29%)
7 (11.67%)
-
10
(16.13%)
-
7 (Strongly
agree)
11
(18.03%)
1
(14.29%)
2 (3.33%)
1
(20.00%)
5 (8.06%)
-
I know car-sharing/bike-
sharing/scooter-sharing
provides good service
(according to the respondents'
previous experience) (answers
only belong to people who have
experience with the service)
1 (Strongly
disagree)
3 (4.92%)
-
7 (11.67%)
1
(20.00%)
6 (9.68%)
-
2
1 (1.64%)
-
2 (3.33%)
2
(40.00%)
4 (6.45%)
-
3
6 (9.84%)
2
(28.57%)
12 (20.00%)
-
6 (9.68%)
1 (5.58%)
4
9
(14.75%)
1
(14.29%)
10 (16.67%)
-
10
(16.13%)
6
(35.29%)
5
13
(21.31%)
3
(42.86%)
17 (28.33%)
1
(20.00%)
13
(20.97%)
6
(35.29%)
6
15
(24.59%)
-
8 (13.33%)
1
(20.00%)
17
(27.42%)
3
(17.65%)
7 (Strongly
agree)
14
(22.95%)
1
(14.29%)
4 (6.67%)
-
6 (9.68%)
1 (5.88%)
I know it is predictable
(according to the respondents'
previous experience) (answers
only belong to people who have
experience with the service)
1 (Strongly
disagree)
1 (1.64%)
-
6 (10.00%)
1
(20.00%)
2 (3.23%)
-
2
3 (4.92%)
1
(14.29%)
3 (5.00%)
-
6 (9.68%)
3
(17.65%)
3
8
(13.11%)
-
9 (15.00%)
2
(40.00%)
7
(11.29%)
6
(35.29%)

402
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
4
10
(16.39%)
3
(42.86%)
14 (23.33%)
1
(20.00%)
18
(29.03%)
6
(35.29%)
5
17
(27.87%)
2
(28.57%)
17 (28.33%)
-
17
(27.42%)
2
(11.76%)
6
10
(16.39%)
-
8 (13.33%)
1
(20.00%)
9
(14.52%)
-
7 (Strongly
agree)
12
(19.67%)
1
(14.29%)
3 (5.00%)
-
3 (4.84%)
-
I know it is trustworthy
(according to the respondents'
previous experience) (answers
only belong to people who have
experience with the service)
1 (Strongly
disagree)
2 (3.28%)
-
6 (10.00%)
1
(20.00%)
5 (8.06%)
2
(11.76%)
2
1 (1.64%)
1
(14.29%)
4 (6.67%)
1
(20.00%)
3 (4.84%)
3
(17.65%)
3
5 (8.20%)
-
8 (13.33%)
-
12
(19.35%)
3
(17.65%)
4
10
(16.39%)
2
(28.57%)
16 (26.67%)
2
(40.00%)
15
(24.19%)
4
(23.53%)
5
14
(22.95%)
2
(28.57%)
11 (18.33%)
-
16
(25.81%)
3
(17.65%)
6
16
(26.23%)
1
(14.29%)
11 (18.33%)
1
(20.00%)
8
(12.90%)
2
(11.76%)
7 (Strongly
agree)
13
(21.31%)
1
(14.29%)
4 (6.67%)
-
3 (4.84%)
-
It would be possible for me to
use it for my regular trips
(According to perceptions of
respondents) (answers only
belong to people who do not
have experience with the
service but are familiar)
1 (Strongly
disagree)
-
19
(21.11%)
-
17
(18.28%)
-
7
(12.96%)
2
-
17
(18.89%)
-
14
(15.05%)
-
9
(16.67%)
3
-
13
(14.44%)
-
18
(19.35%)
-
10
(18.52%)
4
-
17
(18.89%)
-
19
(20.43%)
-
13
(24.07%)
5
-
14
(15.56%)
-
17
(18.28%)
-
10
(18.52%)
6
-
6
(6.67%)
-
3
(3.23%)
-
3 (5.56%)
7 (Strongly
agree)
-
4
(4.44%)
-
5
(5.38%)
-
2 (3.70%)
I am sure that I can choose it
for my regular trips during the
next week (According to
perceptions of respondents)
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
37
(41.11%)
-
43
(46.24%)
-
19
(35.19%)
2
-
12
(13.33%)
-
16
(17.20%)
-
7
(12.96%)
3
-
7
(7.78%)
-
10
(10.75%)
-
3 (5.56%)
4
-
15
(16.67%)
-
9
(9.68%)
-
9
(16.67%)
5
-
12
(13.33%)
-
11
(11.83%)
-
11
(20.37%)
6
-
6
(6.67%)
-
3
(3.23%)
-
3 (5.56%)
7 (Strongly
agree)
-
1
(1.11%)
-
1
(1.08%)
-
2 (3.70%)
Using it would be a useful mode
of transport (According to
perceptions of respondents)
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
12
(13.33%)
-
7
(7.53%)
-
3 (5.56%)
2
-
9
(10.00%)
-
13
(13.98%)
-
7
(12.96%)
3
-
10
(11.11%)
-
8
(8.60%)
-
4 (7.41%)
4
-
22
(24.44%)
-
20
(21.51%)
-
16
(29.63%)
5
-
15
(16.67%)
-
20
(21.51%)
-
13
(24.07%)
6
-
12
(13.33%)
-
13
(13.98%)
-
4 (7.41%)
7 (Strongly
agree)
-
10
(11.11%)
-
12
(12.90%)
-
7
(12.96%)
Using it would help me to
accomplish activities that are
important to me (According to
perceptions of respondents)
1 (Strongly
disagree)
-
26
(28.89%)
-
18
(19.35%)
-
10
(18.52%)
2
-
10
(11.11%)
-
20
(21.51%)
-
12
(22.222%)

403
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
(answers only belong to people
who do not have experience
with the service but are
familiar)
3
-
12
(13.33%)
-
11
(11.83%)
-
3 (5.56%)
4
-
20
(22.22%)
-
17
(18.28%)
-
17
(31.48%)
5
-
12
(13.33%)
-
19
(20.43%)
-
7
(12.96%)
6
-
6
(6.67%)
-
4
(4.30%)
-
5 (9.26%)
7 (Strongly
agree)
-
4
(4.44%)
-
4
(4.30%)
-
-
Learning how to use it would
be easy for me (According to
perceptions of respondents)
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
9
(10.00%)
-
8
(8.60%)
-
3 (5.56%)
2
-
7
(7.78%)
-
9
(9.68%)
-
7
(12.96%)
3
-
4
(4.44%)
-
9
(9.68%)
-
4 (7.41%)
4
-
17
(18.89%)
-
14
(15.05%)
-
12
(22.22%)
5
-
24
(26.67%)
-
20
(21.51%)
-
11
(20.37%)
6
-
20
(22.22%)
-
14
(15.05%)
-
10
(18.52%)
7 (Strongly
agree)
-
9
(10.00%)
-
19
(20.43%)
-
7
(12.96%)
I would find it easy to use
(According to perceptions of
respondents) (answers only
belong to people who do not
have experience with the
service but are familiar)
1 (Strongly
disagree)
-
9
(10.00%)
-
9
(9.68%)
-
3 (5.56%)
2
-
3
(3.33%)
-
7
(7.53%)
-
5 (9.26%)
3
-
5
(5.56%)
-
9
(9.68%)
-
9
(16.67%)
4
-
26
(28.89%)
-
19
(20.43%)
-
13
(24.07%)
5
-
22
(24.44%)
-
15
(16.13%)
-
13
(24.07%)
6
-
18
(20.00%)
-
17
(18.28%)
-
9
(16.67%)
7 (Strongly
agree)
-
7
(7.78%)
-
17
(18.28%)
-
2 (3.70%)
It would be difficult to book it
on the website/app (According
to perceptions of respondents)
(answers only belong to people
who do not have experience
with the service but are
familiar)
1 (Strongly
disagree)
-
30
(33.33%)
-
25
(26.88%)
-
16
(29.63%)
2
-
13
(14.44%)
-
17
(18.28%)
-
7
(12.96%)
3
-
9
(10.00%)
-
10
(10.75%)
-
6
(11.11%)
4
-
15
(16.67%)
-
15
(16.13%)
-
9
(16.67%)
5
-
15
(16.67%)
-
11
(11.83%)
-
11
(20.37%)
6
-
6
(6.67%)
-
11
(11.83%)
-
2 (3.70%)
7 (Strongly
agree)
-
2
(2.22%)
-
4
(4.30%)
-
3 (5.56%)
Using it would be enjoyable
(answers only belong to people
who are at least familiar with
the service)
1 (Strongly
disagree)
-
8
(8.89%)
-
5
(5.38%)
-
3 (5.56%)
2
-
7
(7.78%)
-
8
(8.60%)
-
8
(14.81%)
3
-
9
(10.00%)
-
13
(13.98%)
-
6
(11.11%)
4
-
29
(32.22%)
-
22
(23.66%)
-
15
(27.78%)
5
-
18
(20.00%)
-
21
(22.58%)
-
16
(29.63%)
6
-
13
(14.44%)
-
15
(6.13%)
-
5 (9.26%)
7 (Strongly
agree)
-
6
(6.67%)
-
9
(9.68%)
-
1 (1.85%)
I think that it is
environmentally friendly
1 (Strongly
disagree)
-
8
(8.89%)
-
1
(1.08%)
-
1 (1.85%)

404
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
(answers only belong to people
who are at least familiar with
the service)
2
-
4
(4.44%)
-
1
(1.08%)
-
-
3
-
6
(6.67%)
-
3
(3.23%)
-
4 (7.41%)
4
-
26
(28.89%)
-
9
(9.68%)
-
9
(16.67%)
5
-
16
(17.78%)
-
21
(22.58%)
-
15
(27.78%)
6
-
20
(22.22%)
-
23
(24.73%)
-
19
(35.19%)
7 (Strongly
agree)
-
10
(11.11%)
-
35
(37.63%)
-
6
(11.11%)
I think it provides good service
(According to the respondent's
knowledge) (answers only
belong to people who are at
least familiar with the service)
1 (Strongly
disagree)
-
4
(4.44%)
-
2
(2.15%)
-
-
2
-
3
(3.33%)
-
6
(6.45%)
-
2 (3.70%)
3
-
9
(10.00%)
-
12
(12.90%)
-
6
(11.11%)
4
-
28
(31.11%)
-
19
(20.43%)
-
15
(27.78%)
5
-
27
(30.00%)
-
25
(26.88%)
-
19
(35.19%)
6
-
12
(13.33%)
-
19
(20.43%)
-
12
(22.22%)
7 (Strongly
agree)
-
7
(7.78%)
-
10
(10.75%)
-
-
I think it is predictable
(According to the respondent's
knowledge) (answers only
belong to people who are at
least familiar with the service)
1 (Strongly
disagree)
-
5
(5.56%)
-
3
(3.23%)
-
3 (5.56%)
2
-
4
(4.44%)
-
4
(4.30%)
-
2 (3.70%)
3
-
11
(12.22%)
-
9
(9.68%)
-
8
(14.81%)
4
-
37
(41.11%)
-
27
(29.03%)
-
19
(35.19%)
5
-
20
(22.22%)
-
31
(33.33%)
-
14
(25.93%)
6
-
10
(11.11%)
-
10
(10.75%)
-
8
(14.81%)
7 (Strongly
agree)
-
3
(3.33%)
-
9
(9.68%)
-
-
I think it is trustworthy
(According to the respondent's
knowledge) (answers only
belong to people who are at
least familiar with the service)
1 (Strongly
disagree)
-
6
(6.67%)
-
2
(2.15%)
-
3 (5.56%)
2
-
-
-
7
(7.53%)
-
4 (7.41%)
3
-
11
(12.22%)
-
7
(7.53%)
-
10
(18.52%)
4
-
24
(26.67%)
-
24
(25.81%)
-
18
(33.33%)
5
-
32
(35.56%)
-
26
(27.96%)
-
11
(20.37%)
6
-
10
(11.11%)
-
16
(17.20%)
-
8
(14.81%)
7 (Strongly
agree)
-
7
(7.78%)
-
11
(11.83%)
-
-
The urgent need to reduce
ecological destruction caused
by using the car has been
overestimated
1 (Strongly
disagree)
21
(34.43%)
27
(24.32%)
16 (26.67%)
22
(19.64%)
14
(22.58%)
23
(20.72%)
2
4 (6.56%)
9
(8.11%)
5 (8.33%)
10
(8.93%)
4 (6.45%)
9 (8.11%)
3
4 (6.56%)
19
(17.12%)
4 (6.67%)
13
(11.61%)
4 (6.45%)
7 (6.31%)
4
5 (8.20%)
17
(15.32%)
11 (18.33%)
20
(17.86%)
15
(24.19%)
25
(22.52%)
5
12
(19.67%)
15
(13.51%)
7 (11.67%)
18
(16.07%)
9
(14.52%)
21
(18.92%)
6
3 (4.92%)
10
(9.01%)
8 (13.33%)
33
(11.61%)
9
(14.52%)
15
(13.51%)
7 (Strongly
agree)
12
(19.67%)
14
(12.61%)
9 (15.00%)
16
(14.29%)
7
(11.29%)
11
(9.91%)

405
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
I believe that using a car causes
many environmental problems
1 (Strongly
disagree)
8
(13.11%)
5
(4.50%)
-
1
(0.89%)
4 (6.45%)
5 (4.50%)
2
1 (1.64%)
10
(9.01%)
2 (3.33%)
5
(4.46%)
6 (9.68%)
3 (2.70%)
3
2 (3.28%)
10
(9.01%)
2 (3.33%)
15
(13.39%)
3 (4.84%)
9 (8.11%)
4
6 (9.84%)
25
(22.52%)
10 (16.67%)
16
(14.29%)
14
(22.58%)
23
(20.72%)
5
13
(21.31%)
18
(16.22%)
9 (15.00%)
23
(20.54%)
11
(17.74%)
27
(24.32%)
6
11
(18.03%)
21
(18.92%)
14 (23.33%)
24
(21.43%)
13
(20.97%)
17
(15.32%)
7 (Strongly
agree)
20
(32.79%)
22
(19.82%)
23 (38.33%)
28
(25.00%)
11
(17.74%)
27
(24.32%)
I feel morally obliged to reduce
the environmental impact due
to my travel patterns
1 (Strongly
disagree)
7
(11.48%)
13
(11.71%)
3 (5.00%)
12
(10.71%)
2 (3.23%)
7 (6.31%)
2
3 (4.92%)
9
(8,11%)
3 (5.00%)
13
(11.61%)
5 (8.06%)
9 (8.11%)
3
5 (8.20%)
10
(9.01%)
6 (10.00%)
12
(10.71%)
5 (8.06%)
13
(11.71%)
4
9
(14.75%)
20
(18.02%)
10 (16.67%)
30
(26.79%)
14
(22.58%)
17
(15.32%)
5
7
(11.48%)
23
(20.72%)
15 (25.00%)
16
(14.29%)
12
(19.35%)
30
(27.03%)
6
17
(27.87%)
18
(16.22%)
11 (18.33%)
16
(14.29%)
14
(22.58%)
21
(18.92%)
7 (Strongly
agree)
13
(21.31%)
18
(16.22%)
12 (20.00%)
13
(11.61%)
10
(16.13%)
14
(12.61%)
I would feel guilty if I did not
reduce the environmental
impact of my travel patterns
1 (Strongly
disagree)
9
(14.75%)
14
(12.61%)
5 (8.33%)
14
(12.50%)
5 (8.06%)
6 (5.41%)
2
2 (3.28%)
12
(10.81%)
3 (5.00%)
9
(8.04%)
2 (3.23%)
14
(12.61%)
3
5 (8.20%)
12
(10.81%)
3 (5.00%)
15
(13.39%)
6 (9.68%)
10
(9.01%)
4
8
(13.11%)
21
(18.92%)
14 (23.33%)
26
(23.21%)
15
(24.19%)
18
(16.22%)
5
11
(18.03%)
21
(18.92%)
16 (26.67%)
26
(23.21%)
13
(20.97%)
27
(24.32%)
6
14
(22.95%)
15
(13.51%)
9 (15.00%)
12
(10.71%)
12
(19.35%)
24
(21.62%)
7 (Strongly
agree)
12
(19.67%)
16
(14.41%)
10 (16.67%)
10
(8.93%)
9
(14.52%)
12
(10.81%)
I would feel good if I traveled
more sustainably
1 (Strongly
disagree)
5 (8.20%)
11
(9.91%)
1 (1.67%)
6
(5.36%)
1 (1.61%)
3 (2.70%)
2
-
7
(6.31%)
2 (3.33%)
1
(0.89%)
3 (4.84%)
4 (3.60%)
3
2 (3.28%)
7
(6.31%)
3 (5.00%)
8
(7.14%)
3 (4.84%)
8 (7.21%)
4
12
(19.67%)
21
(18.92%)
10 (16.67%)
24
(21.43%)
14
(22.58%)
18
(16.22%)
5
10
(16.39%)
21
(18.92%)
12 (20.00%)
27
(24.11%)
14
(22.58%)
25
(22.52%)
6
12
(19.67%)
15
(13.51%)
15 (25.00%)
27
(24.11%)
11
(17.74%)
33
(29.73%)
7 (Strongly
agree)
20
(32.79%)
29
(26.13%)
17 (28.33%)
19
(16.96%)
16
(25.81%)
20
(18.02%)
Political issues (green
environmental scale)
1 (Not green)
1 (1.64%)
5
(4.50%)
-
3
(2.68%)
-
2 (1.80%)
2
1 (1.64%)
5
(4.50%)
2 (3.33%)
1
(0.89%)
1 (1.61%)
2 (1.80%)
3
-
6
(5.41%)
1 (1.67%)
6
(5.36%)
4 (6.45%)
6 (5.41%)
4
13
(21.31%)
28
(25.23%)
10 (16.67%)
39
(34.82%)
16
(25.81%)
29
(26.13%)
5
21
(34.43%)
29
(26.13%)
21 (35.00%)
39
(34.82%)
23
(37.10%)
28
(25.23%)
6
13
(21.31%)
24
(21.62%)
18 (30.00%)
13
(11.61%)
10
(16.13%)
32
(28.83%)

406
People's routines and experiences of using shared
mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Users
(n=76)
Non-
users
(n=126)
Users
(n=75)
Non-
users
(n=127)
Users
(n=77)
Non-
users
(n=126)
7 (Very green)
12
(19.67%)
14
(12.61%)
8 (13.33%)
11
(9.82%)
8
(12.90%)
12
(10.81%)
Political issues ("left" or
"right")
Far to the left
9
(14.75%)
3
(2.70%)
9 (15.00%)
4
(3.57%)
4 (6.45%)
7 (6.31%)
Left
8
(13.11%)
15
(13.51%)
12 (20.00%)
16
(14.29%)
6 (9.68%)
18
(16.22%)
Quite left
10
(16.39%)
26
(23.42%)
10 (16.67%)
23
(20.54%)
17
(27.42%)
25
(22.52%)
Neither to the
left nor the right
23
(37.70%)
44
(39.64%)
20 (33.33%)
37
(33.04%)
27
(43.55%)
42
(37.84%)
Quite right
4 (6.56%)
10
(9.01%)
3 (5.00%)
14
(12.50%)
4 (6.45%)
11
(9.91%)
Right
4 (6.56%)
9
(8.11%)
2 (3.33%)
12
(10.71%)
2 (3.23%)
4 (3.60%)
Far to the right
3 (4.92%)
4
(3.60%)
4 (6.67%)
6
(5.36%)
2 (3.23%)
4 (3.60%)
Furthermore, some differences in the routines and daily travel patterns of male and female users
(survey respondents) of each shared transportation service can be seen as listed in Table A12.
Table A12: Differences in the routines and daily travel patterns of male and female users of
each shared transportation service.
User’s routines and experiences of using shared
mobility service
Users of shared Mobility Services
Car-sharing (n=76)
Bike-sharing (n=75)
Scooter-sharing
(n=77)
Males
(n=37)
Females
(n=39)
Males
(n=49)
Females
(n=26)
Males
(n=44)
Females
(n=33)
The incentive to use car-
sharing (or more use)
Availability near my
home/work
1822
19*
-
-
-
-
Reduction in costs
22*
16*
-
-
-
-
More sustainable travel
12*
11*
-
-
-
-
Increased comfort
during travel
2*
10*
-
-
-
-
The convenience of
having it only when
needed
11*
11*
-
-
-
-
Avoiding
responsibilities related
to maintenance and
repairs
8*
8*
-
-
-
-
The incentive to use bike-
sharing/scooter-sharing
(or more use)
Availability near my
home/work
-
-
24*
16*
20*
16*
Reduction in costs
-
-
17*
8*
13*
12*
More sustainable travel
-
-
17*
9*
15*
6*
Increased comfort
during travel
-
-
14*
2*
9*
4*
The convenience of
having it only when
needed
-
-
1623
10*
16*
12*
Avoiding
responsibilities related
to maintenance and
repairs
-
-
8*
6*
9*
9*
Smooth track without
slope
-
-
0 (0.00%)
1*
5*
4*
Departure time (day) that
may cause the use of the
service
Travel on a weekday
morning
6*
17*
27*
16*
25*
19*
Travel on a weekend
morning
15*
10*
20*
11*
22*
17*
22
Respondents could select more than one option, up to three options.
23
Respondents could select more than one option, up to three options.

407
Travel on a weekday
evening
18*
12*
11*
6*
10*
5*
Travel on a weekend
evening
12*
18*
8*
4*
6*
6*
The trip purpose that
may cause the use of the
service
Travel for leisure (e.g.,
vising friends or
shopping)
9
(24.32%)
10
(25.64%)
18
(36.73%)
6
(23.08%)
15
(34.09%)
16
(48.48%)
Travel for non-leisure
(going to work/school)
15
(40.54%)
12
(30.77%)
14
(28.57%)
9
(34.62%)
11
(25.00%)
9
(27.27%)
Both
13
(35.14%)
17
(43.59%)
17
(34.69%)
11
(42.31%)
44
(40.91%)
8
(24.24%)
A4.3 Socio-demographic characteristics of selected users and non-
users of each of the shared mobility services
The socio-demographic characteristics of survey respondents who are users and non-users of
car-sharing, bike-sharing and scooter-sharing services, and their responses to the BWM
questions are selected, listed in Tables A13 to A18 (question set C in surveys 1 to 3). As
mentioned in section 5.4.3 (Chapter 5), after removing pairwise comparisons with unacceptable
consistency ratios, different sample sizes can be obtained and utilized for different levels of the
model.
Table A13: Socio-demographic characteristics of different sets of survey respondents
(selected car-sharing users) (question set C in survey 1), shown in the second row of Table
38.
Socio-demographic factors
Main
criteria set
(n=15)
Trip-related
characteristics
(n= 39)
Car-sharing
characteristics
(n=36)
Availability
and
accessibility
(n=39)
n
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
8
53.33
18
46.15
20
55.56
18
46.15
Female
7
46.67
21
53.85
16
44.44
21
53.85
Age
18-24
0
0.00
1
2.56
2
5.56
1
2.56
25-34
4
26.67
8
20.51
11
30.56
8
20.51
35-44
6
40.00
15
38.46
11
30.56
15
38.46
45-54
1
6.67
11
28.21
10
27.78
11
28.21
55-64
4
26.67
3
7.69
2
5.56
3
7.69
> 64
0
0.00
1
2.56
0
0.00
1
2.56
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
0
0.00
0
0.00
0
0.00
Upper secondary school or
equivalent shorter than three
years
0
0.00
0
0.00
0
0.00
0
0.00
Upper secondary school or
equivalent three years or
more
4
26.67
14
35.90
13
36.11
14
35.90
Post-secondary education,
not college, less than three
years
0
0.00
2
5.13
3
8.33
2
5.13
Post-secondary education,
not college, three years or
more
1
6.67
1
2.56
2
5.56
1
2.56
University less than three
years
0
0.00
1
2.56
3
8.33
1
2.56
University 3 years or more
7
46.67
17
43.59
12
33.33
17
43.59
Degree from postgraduate
studies
3
20.00
4
10.26
3
8.33
4
10.26
Marital status
Single
8
53.33
14
35.90
16
44.44
14
35.90
Married or domestic
partnership
7
46.67
25
64.10
20
55.56
25
64.10
Entrepreneur/freelancer
0
0
1
2.56
3
8.33
1
2.56

408
Business or
professional status
Officer/manager
2
13.33
3
7.69
3
8.33
3
7.69
Clerk/trade employee
7
46.67
21
53.85
19
52.78
21
53.85
Worker
0
0.00
1
2.56
2
5.56
1
2.56
Teacher
1
6.67
2
5.13
0
0.00
2
5.13
Representative
1
6.67
0
0.00
0
0.00
0
0.00
Craftsman / trader / operator
0
0.00
1
2.56
2
5.56
1
2.56
Student
0
0.00
2
5.13
2
5.56
2
5.13
Housewife
0
0.00
3
7.69
1
2.78
3
7.69
Retired
0
0.00
1
2.56
0
0.00
1
2.56
Waiting for first job / never
worked
1
6.67
0
0.00
1
2.78
0
0.00
Unemployed / lost his/her job
0
0.00
1
2.56
1
2.78
1
2.56
Other
3
20.00
3
7.69
2
5.56
3
7.69
Number of people,
including
respondents, living in
the home
One person
4
26.67
4
10.26
3
8.33
4
10.26
Two people
5
33.33
14
35.90
10
27.78
14
35.90
Three people
4
26.67
11
28.21
13
36.11
11
28.21
Four people
1
6.67
7
17.95
6
16.67
7
17.95
Five or more people
1
6.67
3
7.69
4
11.11
3
7.69
Number of drivers,
including
respondents, living in
the home
0
0
0
0
0.00
0
0.00
0
0.00
1
8
53.33
11
28.21
9
25.00
11
28.21
2
5
33.33
19
48.72
16
44.44
19
48.72
More than 2
2
13.33
9
23.08
11
30.56
9
23.08
Presence of children
at home
Yes
7
46.67
19
48.72
18
50.00
19
48.72
No
8
53.33
20
51.28
18
50.00
20
51.28
The age of the
respondent's
child/children
0-3 years old
324
-
7*
-
5*
-
7*
-
4-6 years old
2*
-
3*
-
3*
-
3*
-
7-15 years old
0
0.00
6*
-
5*
-
6*
-
16 years or more
2*
-
7*
-
7*
-
7*
-
Number of cars
available for use in
the respondent's
home
No car
3
20.00
2
5.13
3
8.33
2
5.13
One car
4
26.67
15
38.46
12
33.33
15
38.46
Two cars
7
46.67
19
48.72
18
50.00
19
48.72
Three cars or more
1
6.67
3
7.69
3
8.33
3
7.69
Monthly income of
the respondent after
tax
Up to 500 Euros
0
0.00
2
5.13
2
5.56
2
5.13
501 Euros - 1000 Euros
1
6.67
1
2.56
3
8.33
1
2.56
1001 Euros - 1500 Euros
6
40.00
17
43.59
15
41.67
17
43.59
1501 Euros - 2000 Euros
4
26.67
8
20.51
6
16.67
8
20.51
2001 Euros - 2500 Euros
0
0.00
5
12.82
3
8.33
5
12.82
2501 Euros - 3000 Euros
2
13.33
1
2.56
3
8.33
1
2.56
3001 Euros - 4000 Euros
0
0.00
2
5.13
1
2.78
2
5.13
4001 Euros - 5000 Euros
1
6.67
1
2.56
1
2.78
1
2.56
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
6001 Euros - 10000 Euros
0
0.00
1
2.56
1
2.78
1
2.56
More than 10,001 Euros
1
6.67
1
2.56
1
2.78
1
2.56
Respondent's
household monthly
income after tax
Up to 500 Euros
0
0.00
0
0.00
0
0
0
0.00
501 Euros - 1000 Euros
1
6.67
0
0.00
1
2.78
0
0.00
1001 Euros - 1500 Euros
2
13.33
5
12.82
3
8.33
5
12.82
1501 Euros - 2000 Euros
4
26.67
8
20.51
8
22.22
8
20.51
2001 Euros - 2500 Euros
1
6.67
9
23.08
7
19.44
9
23.08
2501 Euros - 3000 Euros
2
13.33
6
15.38
6
16.67
6
15.38
3001 Euros - 4000 Euros
2
13.33
5
12.82
4
11.11
5
12.82
4001 Euros - 5000 Euros
1
6.67
2
5.13
3
8.33
2
5.13
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
6001 Euros - 10000 Euros
0
0.00
3
7.69
3
8.33
3
7.69
More than 10,001 Euros
2
13.33
1
2.56
1
2.78
1
2.56
How respondents
manage their
expenses with their
current income
Very good
0
0.00
7
17.95
8
22.22
7
17.95
Fairly good
10
66.67
15
38.46
13
36.11
15
38.46
Neither good nor bad
5
33.33
12
30.77
12
33.33
12
30.77
Pretty bad
0
0.00
5
12.82
3
8.33
5
12.82
Very bad
0
0.00
0
0.00
0
0.00
0
0.00
24
Respondents could select more than one option, up to three options.

409
Table A14: Socio-demographic characteristics of different sets of survey respondents (car-
sharing non-users), shown in the third row of Table 38 (question set C in survey 1).
Socio-demographic factors
Main
criteria set
(n=24)
Trip-related
characteristics
(n= 59)
Car-sharing
characteristics
(n=56)
Availability
and
accessibility
(n=59)
n
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
14
58.33
32
54.24
29
51.79
32
54.24
Female
10
41.67
27
45.76
27
48.21
27
45.76
Age
18-24
4
16.67
6
10.17
6
10.71
6
10.17
25-34
4
16.67
9
15.25
9
16.07
9
15.25
35-44
1
4.17
6
10.17
6
10.71
6
10.17
45-54
6
25
11
18.64
15
26.79
11
18.64
55-64
4
16.67
14
23.73
12
21.43
14
23.73
> 64
5
20.83
13
22.03
8
14.29
13
22.03
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
0
0.00
0
0.00
0
0.00
Upper secondary school or
equivalent shorter than three
years
0
0.00
6
10.17
5
8.93
6
10.17
Upper secondary school or
equivalent three years or
more
8
33.33
20
33.90
17
30.36
20
33.90
Post-secondary education,
not college, less than three
years
0
0.00
0
0.00
0
0.00
0
0.00
Post-secondary education,
not college, three years or
more
1
4.17
4
6.78
4
7.14
4
6.78
University less than three
years
1
4.17
1
1.69
2
3.57
1
1.69
University 3 years or more
13
54.17
25
42.37
25
44.64
25
42.37
Degree from postgraduate
studies
1
4.17
3
5.08
3
5.36
3
5.08
Marital status
Single
12
50
26
44.07
21
37.50
26
44.07
Married or domestic
partnership
12
50
33
55.93
35
62.50
33
55.93
Business or
professional status
Entrepreneur/freelancer
0
0.00
1
1.69
2
3.57
1
1.69
Officer/manager
2
8.33
4
6.78
4
7.14
4
6.78
Clerk/trade employee
7
29.17
18
30.51
22
39.29
18
30.51
Worker
1
4.17
1
1.69
2
3.57
1
1.69
Teacher
2
8.33
2
3.39
2
3.57
2
3.39
Representative
0
0.00
2
3.39
0
0.00
2
3.39
Craftsman / trader / operator
0
0.00
1
1.69
0
0.00
1
1.69
Student
4
16.67
8
13.56
6
10.71
8
13.56
Housewife
1
4.17
3
5.08
2
3.57
3
5.08
Retired
6
25.00
14
23.73
10
17.86
14
23.73
Waiting for first job / never
worked
0
0.00
2
3.39
2
3.57
2
3.39
Unemployed / lost his/her job
1
4.17
2
3.39
1
1.79
2
3.39
Other
0
0.00
1
1.69
3
5.36
1
1.69
Number of people,
including
respondents, living in
the home
One person
4
16.67
13
22.03
8
14.29
13
22.03
Two people
9
37.50
21
35.59
16
28.57
21
35.59
Three people
7
29.17
9
15.25
15
26.79
9
15.25
Four people
4
16.67
14
23.73
15
26.79
14
23.73
Five or more people
0
0.00
2
3.39
2
3.57
2
3.39
Number of drivers,
including
respondents, living in
the home
0
0
0.00
0
0.00
0
0.00
0
0.00
1
7
29.17
23
38.98
15
26.79
23
38.98
2
13
54.17
24
40.68
27
48.21
24
40.68
More than 2
4
16.67
12
20.34
14
25.00
12
20.34
Presence of children
at home
Yes
5
20.83
19
32.20
25
44.64
19
32.20
No
19
79.17
40
67.80
31
55.36
40
67.80
0-3 years old
0
0.00
625
-
4*
-
6*
-
4-6 years old
0
0.00
6*
-
3*
-
6*
-
25
Respondents could select more than one option, up to three options.

410
Socio-demographic factors
Main
criteria set
(n=24)
Trip-related
characteristics
(n= 59)
Car-sharing
characteristics
(n=56)
Availability
and
accessibility
(n=59)
n
(%)
n
(%)
n
(%)
n
(%)
The age of the
respondent's
child/children
7-15 years old
4*
-
7*
-
12*
-
7*
-
16 years or more
2*
-
11*
-
6*
-
11*
-
Number of cars
available for use in
respondent's home
No car
1
4.17
4
6.78
3
5.36
4
6.78
One car
10
41.67
30
50.85
24
42.86
30
50.85
Two cars
11
45.83
21
35.59
24
42.86
21
35.59
Three cars or more
2
8.33
4
6.78
5
8.93
4
6.78
Monthly income of
the respondent after
tax
Up to 500 Euros
4
16.67
9
15.25
8
14.29
9
15.25
501 Euros - 1000 Euros
3
12.50
5
8.47
4
7.14
5
8.47
1001 Euros - 1500 Euros
4
16.67
9
15.25
12
21.43
9
15.25
1501 Euros - 2000 Euros
3
12.50
15
25.42
8
14.29
15
25.42
2001 Euros - 2500 Euros
7
29.17
10
16.95
11
19.64
10
16.95
2501 Euros - 3000 Euros
0
0.00
5
8.47
8
14.29
5
8.47
3001 Euros - 4000 Euros
2
8.33
2
3.39
2
3.57
2
3.39
4001 Euros - 5000 Euros
1
4.17
2
3.39
2
3.57
2
3.39
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
2
3.39
1
1.79
2
3.39
Respondent's
household monthly
income after tax
Up to 500 Euros
0
0.00
1
1.69
1
1.79
1
1.69
501 Euros - 1000 Euros
1
4.17
4
6.78
4
7.14
4
6.78
1001 Euros - 1500 Euros
2
8.33
8
13.56
4
7.14
8
13.56
1501 Euros - 2000 Euros
2
8.33
13
22.03
10
17.86
13
22.03
2001 Euros - 2500 Euros
8
33.33
8
13.56
12
21.43
8
13.56
2501 Euros - 3000 Euros
2
8.33
5
8.47
9
16.07
5
8.47
3001 Euros - 4000 Euros
5
20.83
12
20.34
9
16.07
12
20.34
4001 Euros - 5000 Euros
2
8.33
3
5.08
3
5.36
3
5.08
5001 Euros - 6000 Euros
2
8.33
3
5.08
3
5.36
3
5.08
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
2
3.39
1
1.79
2
3.39
How respondents
manage their
expenses with their
current income
Very good
2
8.33
3
5.08
3
5.36
3
5.08
Fairly good
9
37.50
27
45.76
20
35.71
27
45.76
Neither good nor bad
11
45.83
21
35.59
24
42.86
21
35.59
Pretty bad
1
4.17
5
8.47
6
10.71
5
8.47
Very bad
1
4.17
3
5.08
3
5.36
3
5.08
Table A15: Socio-demographic characteristics of different sets of survey respondents (bike-
sharing users), shown in the second row of Table 39 (question set C in survey 2).
Socio-demographic factors
Main
criteria set
(n=15)
Trip-related
characteristics
(n= 38)
Bike-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=38)
n
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
7
46.67
26
68.42
23
62.16
26
68.42
Female
8
53.33
12
31.58
14
37.84
12
31.58
Age
< 18
0
0.00
0
0.00
0
0.00
0
0.00
18-24
0
0.00
0
0.00
0
0.00
0
0.00
25-34
0
0.00
11
28.95
10
27.03
11
28.95
35-44
2
13.33
8
21.05
12
32.43
8
21.05
45-54
6
40.00
8
21.05
9
24.32
8
21.05
55-64
3
20.00
7
18.42
4
10.81
7
18.42
> 64
4
26.67
4
10.53
2
5.41
4
10.53
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
0
0.00
0
0.00
0
0.00
Upper secondary school or
equivalent shorter than three
years
0
0.00
1
2.63
0
0.00
1
2.63
Upper secondary school or
equivalent three years or
more
3
20.00
9
23.68
10
27.03
9
23.68
Post-secondary education,
not college, less than three
years
1
6.67
0
0.00
1
2.70
0
0.00

411
Socio-demographic factors
Main
criteria set
(n=15)
Trip-related
characteristics
(n= 38)
Bike-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=38)
n
(%)
n
(%)
n
(%)
n
(%)
Post-secondary education,
not college, three years or
more
3
20.00
2
5.26
3
8.11
2
5.26
University less than three
years
0
0.00
0
0.00
0
0.00
0
0.00
University 3 years or more
4
26.67
15
39.47
14
37.84
15
39.47
Degree from postgraduate
studies
4
26.67
11
28.95
9
24.32
11
28.95
Marital status
Single
3
20
12
31.58
10
27.03
12
31.58
Married or domestic
partnership
12
80
26
68.42
27
72.97
26
68.42
Business or
professional status
Entrepreneur/freelancer
3
20.00
7
18.42
7
18.92
7
18.42
Officer/manager
0
0.00
4
10.53
5
13.51
4
10.53
Clerk/trade employee
5
33.33
11
28.95
9
24.32
11
28.95
Worker
0
0.00
3
7.89
1
2.70
3
7.89
Teacher
1
6.67
2
5.26
1
2.70
2
5.26
Representative
1
6.67
2
5.26
2
5.41
2
5.26
Craftsman / trader / operator
0
0.00
0
0.00
1
2.70
0
0.00
Student
0
0.00
0
0.00
1
2.70
0
0.00
Housewife
1
6.67
2
5.26
4
10.81
2
5.26
Retired
3
20.00
2
5.26
2
5.41
2
5.26
Waiting for first job / never
worked
0
0.00
0
0.00
0
0.00
0
0.00
Unemployed / lost his/her job
1
6.67
3
7.89
3
8.11
3
7.89
Other
0
0.00
2
5.26
1
2.70
2
5.26
Number of people,
including
respondents, living in
the home
One person
1
6.67
8
21.05
7
18.92
8
21.05
Two people
8
53.33
11
28.95
9
24.32
11
28.95
Three people
4
26.67
8
21.05
8
21.62
8
21.05
Four people
2
13.33
7
18.42
10
27.03
7
18.42
Five or more people
0
0.00
4
10.53
3
8.11
4
10.53
Number of drivers,
including
respondents, living in
the home
0
0
0.00
0
0.00
0
0.00
0
0.00
1
4
26.67
18
47.37
15
40.54
18
47.37
2
10
66.67
12
31.58
15
40.54
12
31.58
More than 2
1
6.67
8
21.05
7
18.92
8
21.05
Presence of children
at home
Yes
6
40
15
39.47
16
43.24
15
39.47
No
9
60
23
60.53
21
56.76
23
60.53
The age of the
respondent's
child/children
0-3 years old
126
-
5*
-
5*
-
5*
-
4-6 years old
0
0.00
2*
-
4*
-
2*
-
7-15 years old
3*
-
7*
-
7*
-
7*
-
16 years or more
2*
-
10*
-
9*
-
10*
-
Number of cars
available for use in
respondent's home
No car
2
13.33
3
7.89
4
10.81
3
7.89
One car
7
46.67
22
57.89
18
48.65
22
57.89
Two cars
6
40.00
11
28.95
13
35.14
11
28.95
Three cars or more
0
0.00
2
5.26
2
5.41
2
5.26
Monthly income of
the respondent after
tax
Up to 500 Euros
3
20.00
2
5.26
4
10.81
2
5.26
501 Euros - 1000 Euros
1
6.67
6
15.79
4
10.81
6
15.79
1001 Euros - 1500 Euros
4
26.67
7
18.42
9
24.32
7
18.42
1501 Euros - 2000 Euros
1
6.67
9
23.68
5
13.51
9
23.68
2001 Euros - 2500 Euros
2
13.33
3
7.89
3
8.11
3
7.89
2501 Euros - 3000 Euros
1
6.67
4
10.53
6
16.22
4
10.53
3001 Euros - 4000 Euros
1
6.67
3
7.89
4
10.81
3
7.89
4001 Euros - 5000 Euros
2
13.33
1
2.63
1
2.70
1
2.63
5001 Euros - 6000 Euros
0
0.00
1
2.63
0
0.00
1
2.63
6001 Euros - 10000 Euros
0
0.00
1
2.63
0
0.00
1
2.63
More than 10,001 Euros
0
0.00
1
2.63
1
2.70
1
2.63
Respondent's
household monthly
income after tax
Up to 500 Euros
2
13.33
2
5.26
2
5.41
2
5.26
501 Euros - 1000 Euros
0
0.00
4
10.53
3
8.11
4
10.53
1001 Euros - 1500 Euros
2
13.33
5
13.16
7
18.92
5
13.16
1501 Euros - 2000 Euros
1
6.67
6
15.79
1
2.70
6
15.79
2001 Euros - 2500 Euros
2
13.33
6
15.79
6
16.22
6
15.79
2501 Euros - 3000 Euros
3
20.00
1
2.63
2
5.41
1
2.63
3001 Euros - 4000 Euros
3
20.00
5
13.16
8
21.62
5
13.16
4001 Euros - 5000 Euros
1
6.67
5
13.16
6
16.22
5
13.16
5001 Euros - 6000 Euros
1
6.67
1
2.63
0
0.00
1
2.63
26
Respondents could select more than one option, up to three options.

412
Socio-demographic factors
Main
criteria set
(n=15)
Trip-related
characteristics
(n= 38)
Bike-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=38)
n
(%)
n
(%)
n
(%)
n
(%)
6001 Euros - 10000 Euros
0
0.00
2
5.26
1
2.70
2
5.26
More than 10,001 Euros
0
0.00
1
2.63
1
2.70
1
2.63
How respondents
manage their
expenses with their
current income
Very good
1
6.67
2
5.26
2
5.41
2
5.26
Fairly good
7
46.67
14
36.84
17
45.95
14
36.84
Neither good nor bad
4
26.67
13
34.21
9
24.32
13
34.21
Pretty bad
2
13.33
8
21.05
8
21.62
8
21.05
Very bad
1
6.67
1
2.63
1
2.70
1
2.63
Table A16: Socio-demographic characteristics of different sets of survey respondents (bike-
sharing non-users) shown in the third row of Table 39 (question set C in survey 2).
Socio-demographic factors
Main
criteria set
(n=32)
Trip-related
characteristics
(n= 69)
Bike-sharing
characteristics
(n=63)
Availability
and
accessibility
(n=69)
n
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
21
65.63
33
47.83
33
52.38
33
47.83
Female
11
34.38
36
52.17
30
47.62
36
52.17
Age
< 18
0
0.00
0
0.00
0
0.00
0
0.00
18-24
1
3.13
2
2.90
2
3.17
2
2.90
25-34
6
18.75
8
11.59
8
12.70
8
11.59
35-44
8
25.00
14
20.29
15
23.81
14
20.29
45-54
7
21.88
21
30.43
15
23.81
21
30.43
55-64
5
15.63
10
14.49
10
15..87
10
14.49
> 64
5
15.63
14
20.29
13
20.63
14
20.29
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
1
1.45
2
3.17
1
1.45
Upper secondary school or
equivalent shorter than three
years
3
9.38
5
7.25
4
6.35
5
7.25
Upper secondary school or
equivalent three years or
more
10
31.25
29
42.03
22
34.92
29
42.03
Post-secondary education,
not college, less than three
years
4
12.50
3
4.35
4
6.35
3
4.35
Post-secondary education,
not college, three years or
more
1
3.13
4
5.80
3
4.76
4
5.80
University less than three
years
0
0.00
6
8.70
4
6.35
6
8.70
University 3 years or more
11
34.38
15
21.74
17
26.98
15
21.74
Degree from postgraduate
studies
3
9.38
6
8.70
7
11.11
6
8.70
Marital status
Single
10
31.25
22
31.88
21
33.33
22
31.88
Married or domestic
partnership
22
68.75
47
68.12
42
66.67
47
68.12
Business or
professional status
Entrepreneur/freelancer
3
9.38
7
10.14
7
11.11
7
10.14
Officer/manager
1
3.13
4
5.80
3
4.76
4
5.80
Clerk/trade employee
13
40.63
23
33.33
20
31.75
23
33.33
Worker
0
0.00
4
5.80
3
4.76
4
5.80
Teacher
2
6.25
3
4.35
2
3.17
3
4.35
Representative
0
0.00
1
1.45
1
1.59
1
1.45
Craftsman / trader / operator
1
3.13
0
0.00
0
0.00
0
0.00
Student
3
9.38
2
2.90
3
4.76
2
2.90
Housewife
1
3.13
5
7.25
3
4.76
5
7.25
Retired
2
6.25
10
14.49
10
15.87
10
14.49
Waiting for first job / never
worked
0
0.00
1
1.45
1
1.59
1
1.45
Unemployed / lost his/her job
5
15.63
8
11.59
9
14.29
8
11.59
Other
1
3.13
1
1.45
1
1.59
1
1.45
Number of people,
including
One person
3
9.38
11
15.94
10
15.87
11
15.94
Two people
13
40.63
27
39.13
27
42.86
27
39.13
Three people
9
28.13
17
24.64
16
25.40
17
24.64

413
Socio-demographic factors
Main
criteria set
(n=32)
Trip-related
characteristics
(n= 69)
Bike-sharing
characteristics
(n=63)
Availability
and
accessibility
(n=69)
n
(%)
n
(%)
n
(%)
n
(%)
respondents, living in
the home
Four people
6
18.75
14
20.29
9
14.29
14
20.29
Five or more people
1
3.13
0
0.00
1
1.59
0
0.00
Number of drivers,
including
respondents, living in
the home
0
0
0.00
0
0.00
1
1.59
0
0.00
1
6
18.75
21
30.43
20
31.75
21
30.43
2
21
65.63
40
57.97
34
53.97
40
57.97
More than 2
5
15.63
8
11.59
8
12.70
8
11.59
Presence of children
at home
Yes
12
37.50
27
39.13
19
30.16
27
39.13
No
20
62.50
42
60.87
44
69.84
42
60.87
The age of the
respondent's
child/children
0-3 years old
327
-
4*
-
2*
-
4*
-
4-6 years old
3*
-
5*
-
5*
-
5*
-
7-15 years old
3*
-
9*
-
5*
-
9*
-
16 years or more
5*
-
13*
-
10*
-
13*
-
Number of cars
available for use in
respondent's home
No car
0
0.00
6
8.70
8
12.70
6
8.70
One car
16
50.00
28
40.58
31
49.21
28
40.58
Two cars
15
46.88
31
44.93
23
36.51
31
44.93
Three cars or more
1
3.13
4
5.80
1
1.59
4
5.80
Monthly income of
the respondent after
tax
Up to 500 Euros
5
15.63
13
18.4
12
19.05
13
18.84
501 Euros - 1000 Euros
4
12.50
7
10.14
6
9.52
7
10.14
1001 Euros - 1500 Euros
4
12.50
10
14.49
10
15.87
10
14.49
1501 Euros - 2000 Euros
10
31.25
18
26.09
15
23.81
18
26.09
2001 Euros - 2500 Euros
3
9.38
10
14.49
8
12.70
10
14.49
2501 Euros - 3000 Euros
2
6.25
3
4.35
5
7.94
3
4.35
3001 Euros - 4000 Euros
4
12.50
5
7.25
6
9.52
5
7.25
4001 Euros - 5000 Euros
0
0.00
1
1.45
1
1.59
1
1.45
5001 Euros - 6000 Euros
0
0.00
1
1.45
0
0.00
1
1.45
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
1
1.45
0
0.00
1
1.45
Respondent's
household monthly
income after tax
Up to 500 Euros
4
12.50
6
8.70
7
11.11
6
8.70
501 Euros - 1000 Euros
1
3.13
3
4.35
3
4.76
3
4.35
1001 Euros - 1500 Euros
4
12.50
12
17.39
10
15.87
12
17.39
1501 Euros - 2000 Euros
4
12.50
8
11.59
7
11.11
8
11.59
2001 Euros - 2500 Euros
5
15.63
12
17.39
7
11.11
12
17.39
2501 Euros - 3000 Euros
5
15.63
10
14.49
14
22.22
10
14.49
3001 Euros - 4000 Euros
7
21.88
12
17.39
12
19.05
12
17.39
4001 Euros - 5000 Euros
1
3.13
2
2.90
2
3.17
2
2.90
5001 Euros - 6000 Euros
1
3.13
3
4.35
1
1.59
3
4.35
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
1
1.45
0
0.00
1
1.45
How respondents
manage their
expenses with their
current income
Very good
1
3.13
7
10.14
3
4.76
7
10.14
Fairly good
16
50.00
31
44.93
31
49.21
31
44.93
Neither good nor bad
9
28.13
21
30.43
17
26.98
21
30.43
Pretty bad
3
9.38
7
10.14
7
11.11
7
10.14
Very bad
3
9.38
3
4.35
5
7.94
3
4.35
Table A17: Socio-demographic characteristics of different sets of survey respondents
(scooter-sharing users), shown in the second row of Table 40 (question set C in survey 3).
Socio-demographic factors
Main
criteria set
(n=13)
Trip-related
characteristics
(n= 42)
Scooter-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=42)
n
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
8
61.54
24
57.14
22
59.46
24
57.14
Female
5
38.46
18
42.86
15
40.54
18
42.86
Age
< 18
0
0.00
0
0.00
0
0.00
0
0.00
18-24
1
7.69
2
4.76
2
5.41
2
4.76
25-34
5
38.46
11
26.19
6
16.22
11
26.19
35-44
2
15.38
4
9.52
9
24.32
4
9.52
45-54
2
15.38
4
9.52
2
5.41
4
9.52
55-64
3
23.08
11
26.19
14
37.84
11
26.19
> 64
0
0.00
10
23.81
4
10.81
10
23.81
27
Respondents could select more than one option, up to three options.

414
Socio-demographic factors
Main
criteria set
(n=13)
Trip-related
characteristics
(n= 42)
Scooter-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=42)
n
(%)
n
(%)
n
(%)
n
(%)
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
1
2.38
0
0.00
1
2.38
Upper secondary school or
equivalent shorter than three
years
1
7.69
3
7.14
4
10.81
3
7.14
Upper secondary school or
equivalent three years or
more
2
15.38
13
30.95
12
32.43
13
30.95
Post-secondary education,
not college, less than three
years
0
0.00
3
7.14
1
2.70
3
7.14
Post-secondary education,
not college, three years or
more
1
7.69
2
4.76
1
2.70
2
4.76
University less than three
years
1
7.69
3
7.14
2
5.41
3
7.14
University 3 years or more
6
46.15
14
33.33
13
35.14
14
33.33
Degree from postgraduate
studies
2
15.38
3
7.14
4
10.81
3
7.14
Marital status
Single
7
53.85
14
33.33
12
32.43
14
33.33
Married or domestic
partnership
6
46.15
28
66.67
25
67.57
28
66.67
Business or
professional status
Entrepreneur/freelancer
0
0.00
2
4.76
1
2.70
2
4.76
Officer/manager
1
7.69
6
14.29
6
16.22
6
14.29
Clerk/trade employee
5
38.46
10
23.81
13
35.14
10
23.81
Worker
1
7.69
1
2.38
2
5.41
1
2.38
Teacher
0
0.00
0
0.00
0
0.00
0
0.00
Representative
0
0.00
1
2.38
0
0.00
1
2.38
Craftsman / trader / operator
1
7.69
3
7.14
2
5.41
3
7.14
Student
3
23.08
5
11.90
4
10.81
5
11.90
Housewife
1
7.69
1
2.38
2
5.41
1
2.38
Retired
0
0.00
11
26.19
6
16.22
11
26.19
Waiting for first job / never
worked
0
0.00
0
0.00
0
0.00
0
0.00
Unemployed / lost his/her job
0
0.00
0
0.00
0
0.00
0
0.00
Other
1
7.69
2
4.76
1
2.70
2
4.76
Number of people,
including
respondents, living in
the home
One person
4
30.77
11
26.19
7
18.92
11
26.19
Two people
4
30.77
15
35.71
11
29.73
15
35.71
Three people
2
15.38
9
21.43
13
35.14
9
21.43
Four people
2
15.38
6
14.29
5
13.51
6
14.29
Five or more people
1
7.69
1
2.38
1
2.70
1
2.38
Number of drivers,
including
respondents, living in
the home
0
1
7.69
5
11.90
2
5.41
5
11.90
1
5
38.46
15
35.71
11
29.73
15
35.71
2
4
30.77
11
26.19
12
32.43
11
26.19
More than 2
3
23.08
11
26.19
12
32.43
11
26.19
Presence of children
at home
Yes
2
15.38
11
26.19
12
32.43
11
26.19
No
11
84.62
31
73.81
25
67.57
31
73.81
The age of the
respondent's
child/children
0-3 years old
0
0.00
0
0.00
0
0.00
0
0.00
4-6 years old
0
0.00
0
0.00
128
-
0
0.00
7-15 years old
0
0.00
2*
-
1*
-
2*
-
16 years or more
2*
-
10*
-
6*
-
10*
-
Number of cars
available for use in
respondent's home
No car
0
0.00
3
7.14
1
2.70
3
7.14
One car
10
76.92
21
50.00
17
45.95
21
50.00
Two cars
2
15.38
14
33.33
16
43.24
14
33.33
Three cars or more
1
7.69
4
9.52
3
8.11
4
9.52
Monthly income of
the respondent after
tax
Up to 500 Euros
2
15.38
5
11.90
4
10.81
5
11.90
501 Euros - 1000 Euros
1
7.69
3
7.14
1
2.70
3
7.14
1001 Euros - 1500 Euros
1
7.69
10
23.81
5
13.51
10
23.81
1501 Euros - 2000 Euros
7
53.85
9
21.43
12
32.43
9
21.43
2001 Euros - 2500 Euros
1
7.69
6
14.29
3
8.11
6
14.29
2501 Euros - 3000 Euros
0
0.00
5
11.90
6
16.22
5
11.90
3001 Euros - 4000 Euros
0
0.00
3
7.14
3
8.11
3
7.14
4001 Euros - 5000 Euros
1
7.69
1
2.38
3
8.11
1
2.38
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
28
Respondents could select more than one option, up to three options.

415
Socio-demographic factors
Main
criteria set
(n=13)
Trip-related
characteristics
(n= 42)
Scooter-sharing
characteristics
(n=37)
Availability
and
accessibility
(n=42)
n
(%)
n
(%)
n
(%)
n
(%)
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
0
0.00
0
0.00
0
0.00
Respondent's
household monthly
income after tax
Up to 500 Euros
1
7.69
1
2.38
1
2.70
1
2.38
501 Euros - 1000 Euros
0
0.00
1
2.38
0
0.00
1
2.38
1001 Euros - 1500 Euros
1
7.69
8
19.05
3
8.11
8
19.05
1501 Euros - 2000 Euros
5
38.46
8
19.05
7
18.92
8
19.05
2001 Euros - 2500 Euros
0
0.00
8
19.05
6
16.22
8
19.05
2501 Euros - 3000 Euros
1
7.69
7
16.67
5
13.51
7
16.67
3001 Euros - 4000 Euros
1
7.69
3
7.14
6
16.22
3
7.14
4001 Euros - 5000 Euros
2
15.38
3
7.14
4
10.81
3
7.14
5001 Euros - 6000 Euros
0
0.00
1
2.38
2
5.41
1
2.38
6001 Euros - 10000 Euros
2
15.38
2
4.76
3
8.11
2
4.76
More than 10,001 Euros
0
0.00
0
0.00
0
0.00
0
0.00
How respondents
manage their
expenses with their
current income
Very good
2
15.38
5
11.90
8
21.62
5
11.90
Fairly good
4
30.77
13
30.95
16
43.24
13
30.95
Neither good nor bad
4
30.77
17
40.48
9
24.32
17
40.48
Pretty bad
2
15.38
5
11.90
3
8.11
5
11.90
Very bad
1
7.69
2
4.76
1
2.70
2
4.76
Table A18: Socio-demographic characteristics of different sets of survey respondents
(scooter-sharing non-users) shown in the third row of Table 40 (question set C in survey 3).
Socio-demographic factors
Main
criteria set
(n=24)
Trip-related
characteristics
(n= 66)
Scooter-sharing
characteristics
(n=48)
Availability
and
accessibility
(n=66)
N
(%)
n
(%)
n
(%)
n
(%)
Gender
Male
11
45.83
30
45.45
24
50.00
30
45.45
Female
13
54.17
36
54.55
24
50.00
36
54.55
Age
< 18
0
0.00
0
0.00
0
0.00
0
0.00
18-24
1
4.17
1
1.52
1
2.08
1
1.52
25-34
2
8.33
11
16.67
6
12.50
11
16.67
35-44
6
25.00
18
27.77
12
25.00
18
27.77
45-54
11
45.83
22
33.33
18
37.50
22
33.33
55-64
3
12.50
10
15.15
10
20.83
10
15.15
> 64
1
4.17
4
6.06
1
2.08
4
6.06
Education level
Not completed primary
school
0
0.00
0
0.00
0
0.00
0
0.00
Elementary school
0
0.00
0
0.00
0
0.00
0
0.00
Upper secondary school or
equivalent shorter than three
years
0
0.00
5
7.58
2
4..17
5
7.58
Upper secondary school or
equivalent three years or
more
7
29.17
18
27.27
17
35.42
18
27.27
Post-secondary education,
not college, less than three
years
0
0.00
2
3.03
2
4.17
2
3.03
Post-secondary education,
not college, three years or
more
3
12.50
6
9.09
4
8.33
6
9.09
University less than three
years
3
12.50
1
1.52
3
6.25
1
1.52
University 3 years or more
11
45.83
30
45.45
17
35.42
30
45.45
Degree from postgraduate
studies
0
0.00
4
6.06
3
6.25
4
6.06
Marital status
Single
9
37.50
26
39.39
19
39.58
26
39.39
Married or domestic
partnership
15
62.50
40
60.61
29
60.42
40
60.61
Business or
professional status
Entrepreneur/freelancer
1
4.17
4
6.06
5
10.42
4
6.06
Officer/manager
0
0.00
7
10.61
6
12.50
7
10.61
Clerk/trade employee
12
50.00
29
43.94
22
45.83
29
43.94
Worker
1
4.17
7
10.61
3
6.25
7
10.61
Teacher
1
4.17
3
4.55
0
0.00
3
4.55
Representative
0
0.00
0
0.00
0
0.00
0
0.00
Craftsman / trader / operator
0
0.00
0
0.00
0
0.00
0
0.00

416
Socio-demographic factors
Main
criteria set
(n=24)
Trip-related
characteristics
(n= 66)
Scooter-sharing
characteristics
(n=48)
Availability
and
accessibility
(n=66)
N
(%)
n
(%)
n
(%)
n
(%)
Student
2
8.33
3
4.55
4
8.33
3
4.55
Housewife
4
16.67
3
4.55
3
6.25
3
4.55
Retired
1
4.17
4
6.06
1
2.08
4
6.06
Waiting for first job / never
worked
0
0.00
1
1.52
0
0.00
1
1.52
Unemployed / lost his/her job
1
4.17
2
3.03
2
4.17
2
3.03
Other
1
4.17
3
4.55
2
4.17
3
4.55
Number of people,
including
respondents, living in
the home
One person
8
33.33
13
19.70
10
20.83
13
19.70
Two people
4
16.67
25
37.88
16
33.33
25
37.88
Three people
6
25.00
12
18.18
9
18.75
12
18.18
Four people
6
25.00
15
22.73
12
25.00
15
22.73
Five or more people
0
0.00
1
1.52
1
2.08
1
1.52
Number of drivers,
including
respondents, living in
the home
0
3
12.50
5
7.58
4
8.33
5
7.58
1
6
25.00
16
24.24
12
25.00
16
24.24
2
9
37.50
38
57.58
24
50.00
38
57.58
More than 2
6
25.00
7
10.61
8
16.67
7
10.61
Presence of children
at home
Yes
9
37.50
23
34.85
19
39.58
23
34.85
No
15
62.50
43
65.15
29
60.42
43
65.15
The age of the
respondent's
child/children
0-3 years old
12
-
229
-
2*
-
2*
-
4-6 years old
1*
-
3*
-
1*
-
3*
-
7-15 years old
4*
-
11*
-
11*
-
11*
-
16 years or more
5*
-
10*
-
9*
-
10*
-
Number of cars
available for use in
respondent's home
No car
2
8.33
8
12.12
6
12.50
8
12.12
One car
11
45.83
30
45.45
20
41.67
30
45.45
Two cars
8
33.33
25
37.88
20
41.67
25
37.88
Three cars or more
3
12.50
3
4.55
2
4.17
3
4.55
Monthly income of
the respondent after
tax
Up to 500 Euros
6
25.00
8
12.12
6
12.50
8
12.12
501 Euros - 1000 Euros
2
8.33
5
7.58
4
8.33
5
7.58
1001 Euros - 1500 Euros
3
12.50
13
19.70
8
16.67
13
19.70
1501 Euros - 2000 Euros
8
33.33
22
33.33
17
35.42
22
33.33
2001 Euros - 2500 Euros
2
8.33
7
10.61
3
6.25
7
10.61
2501 Euros - 3000 Euros
2
8.33
8
12.12
5
10.42
8
12.12
3001 Euros - 4000 Euros
0
0.00
2
3.03
4
8.33
2
3.03
4001 Euros - 5000 Euros
1
4.17
1
1.52
1
2.08
1
1.52
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
6001 Euros - 10000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
More than 10,001 Euros
0
0.00
0
0.00
0
0.00
0
0.00
Respondent's
household monthly
income after tax
Up to 500 Euros
3
12.50
4
6.06
2
4.17
4
6.06
501 Euros - 1000 Euros
0
0.00
2
3.03
1
2.08
2
3.03
1001 Euros - 1500 Euros
1
4.17
9
13.64
4
8.33
9
13.64
1501 Euros - 2000 Euros
6
25.00
14
21.21
13
27.08
14
21.21
2001 Euros - 2500 Euros
3
12.50
6
9.09
3
6.25
6
9.09
2501 Euros - 3000 Euros
6
25.00
12
18.18
9
18.75
12
18.18
3001 Euros - 4000 Euros
3
12.50
11
16.67
10
20.83
11
16.67
4001 Euros - 5000 Euros
2
8.33
6
9.09
5
10.42
6
9.09
5001 Euros - 6000 Euros
0
0.00
0
0.00
0
0.00
0
0.00
6001 Euros - 10000 Euros
0
0.00
2
3.03
1
2.08
2
3.03
More than 10,001 Euros
0
0.00
0
0.00
0
0.00
0
0.00
How respondents
manage their
expenses with their
current income
Very good
0
0.00
2
3.03
2
4.17
2
3.03
Fairly good
8
33.33
24
36.36
18
37.50
24
36.36
Neither good nor bad
11
45.83
25
37.88
23
47.92
25
37.88
Pretty bad
3
12.50
13
19.70
3
6.25
13
19.70
Very bad
2
8.33
2
3.03
2
4.17
2
3.03
A4.4 Perspectives of whole operators and members of the
government regarding some of the travel routines of users of each
of the shared transportation services
It is important to figure out the opinions of operators (related to each shared mobility service)
and government members about some of the travel routines of users of each shared mobility
29
Respondents could select more than one option, up to three options.

417
service, shown in Table A19 (question set D in surveys 4 to 6). This helps to determine the
gaps between the views of operators and government members about the travel routine of users
of each shared mobility and what users stated about it.
Table A19: Operators’ (associated with each shared mobility service) and government
members’ views on some of the travel routines of users of each shared mobility service
(question set D in surveys 4 to 6).
People's routines and experiences of
using shared mobility service
Shared mobility services
Car-sharing
Bike-sharing
Scooter-sharing
Operators
Government
members
Operators
Government
members
Operators
Government
members
Travel distance
that may cause
the use of the
service
Short-distance
travel (less than 5
km)
-
1 (25.00%)
2 (100.00%)
4 (80.00%)
1
(100.00%)
3 (100.00%)
Long-distance
travel (5 km or
more)
1
(33.33%)
2 (50.00%)
-
-
-
-
Both
2
(66.67%)
1 (25.00%)
-
1 (20.00%)
-
-
Travel time that
may cause the
use of the service
Travel less than
30 min
-
2 (50.00%)
2 (100.00%)
4 (80.00%)
1
(100.00%)
3 (100.00%)
Travel 30 min or
more
1
(33.33%)
1 (25.00%)
-
-
-
-
Both
2
(66.67%)
1 (25.00%)
-
1 (20.00%)
-
-
Departure time
(hour) that may
cause the use of
the service
Travel during
peak hours
1 (0.33%)
1 (25.00%)
-
1 (20.00%)
-
1 (33.33%)
Travel during
off-peak hours
1 (0.33%)
-
-
-
-
-
Both
1 (0.33%)
3 (75.00%)
2 (100.00%)
4 (80.00%)
1
(100.00%)
2 (66.67%)
Departure time
(day) that may
cause the use of
the service
Travel on a
weekday
morning
130
9*
2*
5*
1*
3*
Travel on a
weekend
morning
1*
1*
0 (0.00%)
1*
0 (0.00%)
1*
Travel on a
weekday evening
0 (0.00%)
2*
2*
3*
1*
2*
Travel on a
weekend evening
2*
3*
1*
0 (0.00%)
1*
0 (0.00%)
The trip purpose
that may cause
the use of the
service
Travel for leisure
(e.g., vising
friends or
shopping)
1 (0.33%)
2 (50.00%)
-
-
-
-
Travel for non-
leisure (going to
work/school)
-
-
-
4 (80.00%)
-
1 (33.33%)
Both
2
(66.67%)
2 (50.00%)
2 (100.00%)
1 (20.00%)
1
(100.00%)
2 (66.67%)
30
Respondents could select more than one option, up to three options.