Figure 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Public transport has become one of the major transport options, especially when it comes to reducing motorized individual transport and achieving sustainability while reducing emissions, noise and so on. The use of public transport data has evolved and rapidly improved over the past decades. Indeed, the availability of data from different sources,...
Context in source publication
Context 1
... possible foundation for developing the intended framework may be found in a basic three-layer model from IM; see, e.g., [17,205]. The framework is depicted in Figure 1. The basic, but often neglected issue is that not only the available data are explored, but that the definition of appropriate functionality requirements is privileged. ...
Citations
... This focus on selecting nearby airports showed that always myopically choosing the closest or largest nearby airport can result in less reliable itineraries. Ge et al. (2021) highlight the importance that multimodal itinerary applications have in integrating all available mobility services and data sources into one framework to support the traveler in their decision-making process. Bucher et al. (2017) propose to precompute candidate stops for the first and the last mile in a preprocessing step of the actual routing. ...
Urban travelers today are seeking increasingly more information to plan their optimal trip, based on additional factors other than scheduled departure times. Still, some route planning applications provide a simple approach with a few parameter settings (e.g. to minimize travel time between two specific places at a certain time) and without any multimodal solutions. Our approach provides travelers with a set of non-dominated nearby stops that presents a number of traveler preferences in an easily comprehensible and quickly calculable manner. We display first and last-mile stops that fall on a Pareto front based on multiple criteria such as travel time, number of transfers, and frequency of service. Our algorithm combines stop and route-based information to quickly present the traveler with numerous nearby quality options for their itinerary decision making. We expand this algorithm to include multimodal itineraries with the incorporation of free-floating scooters to investigate the change in stop and itinerary characteristics. We then analyze the results on the star-shaped public transportation network of Göttingen, Germany, to show what advantages stops on the Pareto front have as well as demonstrate the increased effect on frequency and service lines when incorporating a broadened multimodal approach.
... And while this is a challenge facing the broader computer vision community, it is further exacerbated in domain-specific applications like transportation systems (Dilek and Dener 2023). In the specific context of public transit, the need to acquire external technology and talent for these tasks of data acquisition and integration is often cost-prohibitive, hindering the ability to benefit from fusing the abundant traditional and new-found data sources (Ge et al. 2021). The authors of this paper see immense value in incorporating the domain knowledge of relevant data sources to create a streamlined framework for image data acquisition, labeling, and vision model training combining roadside imagery with transit data sources. ...
Accurate travel time estimation is paramount for providing transit users with reliable schedules and dependable real-time information. This work is the first to utilize roadside urban imagery to aid transit agencies and practitioners in improving travel time prediction. We propose and evaluate an end-to-end framework integrating traditional transit data sources with a roadside camera for automated image data acquisition, labeling, and model training to predict transit travel times across a segment of interest. First, we show how the General Transit Feed Specification real-time data can be utilized as an efficient activation mechanism for a roadside camera unit monitoring a segment of interest. Second, automated vehicle location data is utilized to generate ground truth labels for the acquired images based on the observed transit travel time percentiles across the camera-monitored segment during the time of image acquisition. Finally, the generated labeled image dataset is used to train and thoroughly evaluate a Vision Transformer (ViT) model to predict a discrete transit travel time range (band). The results of this exploratory study illustrate that the ViT model is able to learn image features and contents that best help it deduce the expected travel time range with an average validation accuracy ranging between 80 and 85%. We assess the interpretability of the ViT model’s predictions and showcase how this discrete travel time band prediction can subsequently improve continuous transit travel time estimation. The workflow and results presented in this study provide an end-to-end, scalable, automated, and highly efficient approach for integrating traditional transit data sources and roadside imagery to improve the estimation of transit travel duration. This work also demonstrates the added value of incorporating real-time information from computer-vision sources, which are becoming increasingly accessible and can have major implications for improving transit operations and passenger real-time information.
... Public transport APIs: Public transport APIs often follow open standards, such as the General Transit Feed Specification (GTFS), which defines a standard format for sharing information about public transport timetables, routes, and stops. These APIs often use REST to access real-time data, such as the location of vehicles [17]. 3. ...
To further evolve urban quality of life, this paper explores the potential of crowdsensing and crowdsourcing in the context of smart cities. To aid urban planners and residents in understanding the nuances of day-to-day urban dynamics, we actively pursue the improvement of data visualisation tools that can adapt to changing conditions. An architecture was created and implemented that ensures secure and easy connectivity between various sources, such as a network of Internet of Things (IoT) devices, to merge with crowdsensing data and use them efficiently. In addition, we expanded the scope of our study to include the development of mobile and online applications, emphasizing the integration of autonomous and geo-surveillance. The main findings highlight the importance of sensor data in urban knowledge. Their incorporation via Tepresentational State Transfer (REST) Application Programming Interface (APIs) improves data access and informed decision-making, and dynamic data visualisation provides better insights. The geofencing of the application encourages community participation in urban planning and resource allocation, supporting sustainable urban innovation.
... Transit systems are the backbone of the transportation infrastructure, specifically in large cities (Chakroborty et al. 2001). The need to meet mobility, safety, and environmental objectives place greater demands on public transit systems (Covic and Voß 2019;Diab et al. 2021;Ge et al. 2021;Aemmer et al. 2022). Existing public transit systems must expand service regions, increase service frequency, and increase the efficiency to serve the growing needs of the public (Aemmer et al. 2022;Webb et al. 2020). ...
This study evaluates the transferability of the calibrated parameters for mobility performance of transit signal priority (TSP) in a microscopic simulation environment. The analysis is based on two transit corridors in Florida. Two microscopic simulation VISSIM models, a base model, and a TSP model are developed for each corridor. The simulation models are calibrated to represent field conditions. Three driving behavior parameters that significantly affect the simulation results are identified and selected for the transferability study. A genetic algorithm technique is used to obtain an improved value for each of the three parameters for both transit corridors. Calibrated parameters obtained from the first study corridor, which maximize the correlation between simulated and field travel time, are used to estimate the second study corridor’s travel time and compare the results to parameters optimized specifically for the second study corridor. The study uses the application-based and estimation-based approaches for the analysis. Overall, the TSP model parameter results are generally transferable between the two transit corridors. A percentage change of 9.25 and 18.50% are observed for two of the parameters between two TSP corridors which indicates that these two parameters are transferable. On the other hand, one of the parameters with a high percentage change value of 23.80% between the two TSP corridors are not transferable. The findings of this study may present key considerations for transportation agencies and practitioners when planning future TSP deployments.
... Data and technology seem to be available to enhance bus bridging and bus bunching decisions. The inclusion of these data into current databases available online (see, e.g., [58,71]) and a comprehensive connection to generative AI tools to further improve decision-making abilities of individual users of public transport seems a most suitable area for future research, too. ...
Regarding tools and systems from artificial intelligence (AI), chat-based ones from the area of generative AI have become a major focus regarding media coverage. ChatGPT and occasionally other systems (such as those from Microsoft and Google) are discussed with hundreds if not thousands of academic papers as well as newspaper articles. While various areas have considerably gone into this discussion, transportation and logistics has not yet come that far. In this paper, we explore the use of generative AI tools within this domain. More specifically, we focus on a topic related to sustainable passenger transportation, that is, the handling of disturbances in public transport when it comes to bus bunching and bus bridging. The first of these concepts is related to analyzing situations where we observe two or more buses of the same line following close to each other without being planned deliberately and the second is related to the case where buses are used to replace broken connections in other systems, such as subways. Generative AI tools seem to be able to provide meaningful entries and a lot of food for thought while the academic use may still be classified as limited.
... The popularity of social media platforms such as Twitter allows researchers to combine the benefits of sensing via treating people as sensors (Goodchild 2007), with the benefits of capturing the passengers' perspectives (Ge et al. 2021). This opportunity has recently been highlighted by Haghighi et al. (2018), Rahimi et al. (2020), Lock and Pettit (2020) and Mishra and Panda (2022). ...
Responsive management of public transport nodes relies on constant monitoring of service quality. Social media content provides a unique opportunity to detect and monitor events impacting service quality in these nodes, as well as predicting future occurrences of such events. However, the confined geographic area of transport nodes exacerbates the sparsity of available feeds, raising two major challenges: limited observations—leading to biased models—and the asynchronous nature of observations—impeding the detection of causal patterns. Thus, this paper proposes a framework based on a multivariate Hawkes point process and sentiment analysis. The multivariate Hawkes point process allows effective modelling of events without making them discrete, hence it is less affected by data sparsity compared to time series models while enabling the prediction of how certain events can trigger future events. Besides, the extracted sentiments from social media feeds provide additional knowledge about passengers’ perception and thus, are used in our approach to strengthening the model. Experiments on a real-world dataset demonstrate the effectiveness of the model in identifying causal relations over the public transport nodes. They also show the efficacy of the proposed solution in predicting events over the limited context compared to state-of-the-art approaches.
... We also used data from GitHub, an open-source database (McGovern 2016; NYTimes COVID-19 data bot and Sun 2020). Using data drawn from widely varying sources in order to analyze transit crowding standards is an increasingly common transportation research method (Ge et al. 2021). ...
The COVID-19 pandemic dramatically affected public transit systems around the globe. Because transit systems typically move many people closely together on buses and trains, public health guidance demanded that riders should keep a distance of about two meters to others changed the definition of “crowding” on transit in 2020. Accordingly, this research examines how U.S. public transit agencies responded to public health guidance that directly conflicted with their business model. To do this, we examined published crowding standards before the COVID-19 pandemic for a representative sample of 200 transit systems, including whether they started or changed their published standards during the pandemic, as well as the reasons whether agencies publicize such standards at all. We present both descriptive statistics and regression model results to shed light on the factors associated with agency crowding standards. We find that 56% of the agencies surveyed published crowding standards before the pandemic, while only 46% published COVID-19-specific crowding standards. Regression analyses suggest that larger agencies were more likely to publish crowding standards before and during the COVID-19 pandemic, likely because they are more apt to experience crowding. Pandemic-specific crowding standards, by contrast, were associated with a more complex set of factors. We conclude that the relative lack of pandemic standards reflects the uncertainty and fluidity of the public health crisis, inconsistent and at times conflicting with the guidance from public health officials, and, in the U.S., a lack national or transit industry consensus on appropriate crowding standards during the first year of the pandemic.
... It's up to the task at hand to choose the method of analysis to apply. For unsupervised algorithms, clustering and source signal separation are typical uses; for supervised algorithms, classification and prediction/regression are prominent applications Semi-supervised techniques are employed when big unlabeled datasets cannot be handled by standard supervised algorithms (Ge et al., 2021). Structure similarities between labeled and unlabeled data are used in semi-supervised learning approaches in order to broaden the scope of the functional mapping across large datasets. ...
... Data cubes, histogram binning, and hierarchical aggregation are a few examples to consider. Real-time visualizations and user interactivity are the emphasis of the interactive visualization approach (Ge et al., 2021). ...
Using big data in supply chain management (SCM) has the potential to have a significant impact on the industry in general and international transportation in particular. Big data have a direct influence on transportation capacity in future cities. There has been a significant increase in urbanization over the last decade in which one in three people will live in an urban area by 2050. An updated transportation infrastructure is essential to keep up with the present flow of goods, while also limiting its impact on the environment and human health and this is likely to be achieved using big data analytics technique.
To overcome this problem, smart cities are becoming more popular. With the use of information and communication technology (ICT), a smart city aims to address public concerns in an inclusive, municipally-based partnership. A big data transportations system may be built using the superstructure of a smart city. A good way to define it is the modeling and analysis of urban transportation and distribution networks using enormous data sets created by GPS, mobile phones, and transactional data from company activities.
Big Data analytics may be used in public transportation to better understand how people go about the city. A better understanding of passengers’ travel patterns might help transportation providers make better judgments regarding service quality. People who travel by automobile on a regular basis may now be predicted based on the triangulation of mobile phone data from millions of anonymous users. Local and national polls may demonstrate the paradigm’s applicability. To compute the time it takes for passengers to board and exit trains, Metro and iBus vehicle position data may be combined with information from smart cards. Big Data analytics for traffic management may benefit from these findings.
... The data were collected through a relatively simple questionnaire-based survey. In recent years, a wide range of new public transport data have become available from different sources, including big data and remote sensing data (Ge et al. 2021). While georeferenced data tracked by automatic vehicle location systems have great potential for providing the accurate, real-time information on movements of people and freight, automatic fare collection systems can also generate detailed data on travel origin and destination, allowing to analyze the travel behavior and demand. ...
Informal public transport has been growing rapidly in many developing countries. Because urban infrastructure development tends to lag behind rapid population growth, informal public transport often meets the growing gap between demand and supply in urban mobility. Despite the rich literature primarily focused on formal transport modes, the informal transport sector is relatively unknown. The paper analyzes the demand behavior in the “informal” minibus sector in Antananarivo, Madagascar, taking advantage of a recent user survey of thousands of people. It is found that the demand for informal public transport is generally inelastic. Essentially, people have no other choice but to use this kind of public transport. While the time elasticity is estimated at − 0.02 to − 0.05, the price elasticity is − 0.05 to − 0.06 for short-distance travelers, who may have alternative choices, such as motorcycle taxi or walking. Unlike formal public transportation, the demand also increases with income. Regardless of the income level, everyone uses minibuses. The estimated demand functions indicate that people prefer safety and more flexibility in transit. The paper shows that combining these improvements and fare adjustments, the informal transport sector can contribute to increasing people’s mobility and reducing traffic congestion in the city.
... A burgeoning wealth of transit data sources such as automated fare collection (AFC), automated passenger count (APC), AVL data, smart cards, phones, and social media websites are documented in recent literature (Ge et al. 2021;Li et al. 2018;Lu et al. 2021). In particular, AFC, APC, and AVL data provide high resolution, targeted insights for measuring transit performance. ...
... Several works have distinguished which problems these datasets may support for the collecting agency, that are not better served by typical automated data (Diaz et al. 2021;Jevinger and Persson 2019); for example, using social media posts as a measure of communication between transit agencies and the public, or using cellular routing information to understand passenger destinations. Last, although many automated collection methods have replaced the need for traditional surveys, there may still be value in using surveys as a complement to these "big data" sources (Ge et al. 2021). In particular, census data is essential for observing socio-demographic trends. ...
... Public-facing GTFS-RT feeds provide a standardized format for repackaging AVL data, from which schedule padding and other system characteristics can be reliably ascertained (Ge et al. 2021;Wessel and Widener 2017). Many agencies currently provide this data through application programming interfaces (APIs), from which users may request and receive unified results. ...
This paper presents a method for extracting transit performance metrics from a General Transit Feed Specification’s Real-Time (GTFS-RT) component and aggregating them to roadway segments. A framework is then used to analyze this data in terms of consistent, predictable delays (systematic delays) and random variation on a segment-by-segment basis (stochastic delays). All methods and datasets used are generalizable to transit systems which report vehicle locations in terms of GTFS-RT parameters. This provides a network-wide screening tool that can be used to determine locations where reactive treatments (e.g., schedule padding) or proactive infrastructural changes (e.g., bus-only lanes, transit signal priority) may be effective at improving efficiency and reliability. To demonstrate this framework, a case study is performed regarding one year of GTFS-RT data retrieved from the King County Metro bus network in Seattle, Washington. Stochastic and systematic delays were calculated and assigned to segments in the network, providing insight to spatial trends in reliability and efficiency. Findings for the study network suggest that high-pace segments create an opportunity for large, stochastic speedups, while the network as a whole may carry excessive schedule padding. In addition to the static analysis discussed in this paper, an online interactive visualization tool was developed to display ongoing performance measures in the case study region. All code is open-source to encourage additional generalizable work on the GTFS-RT standard.
