Content uploaded by Simon Elias Bibri
Author content
All content in this area was uploaded by Simon Elias Bibri on Nov 03, 2023
Content may be subject to copyright.
Content uploaded by Simon Elias Bibri
Author content
All content in this area was uploaded by Simon Elias Bibri on Nov 03, 2023
Content may be subject to copyright.
Review
Smarter eco-cities and their leading-edge artificial intelligence of
things solutions for environmental sustainability: A comprehensive
systematic review
Simon Elias Bibri
a
,
*
, John Krogstie
b
, Amin Kaboli
c
, Alexandre Alahi
a
a
School of Architecture, Civil and Environmental Engineering (ENAC), Civil Engineering Institute (IIC), Visual Intelligence for Transportation (VITA), Swiss
Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland
b
Department of Computer Science, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
c
School of Engineering, Institute of Mechanical Engineering, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland
article info
Article history:
Received 7 May 2023
Received in revised form
28 September 2023
Accepted 28 September 2023
Keywords:
Smarter eco-cities
Smart eco-cities
Smart cities
Artificial intelligence
Artificial intelligence of things
Machine learning
Environmental sustainability
Climate change
abstract
The recent advancements made in the realms of Artificial Intelligence (AI) and Artificial Intelligence of
Things (AIoT) have unveiled transformative prospects and opportunities to enhance and optimize the
environmental performance and efficiency of smart cities. These strides have, in turn, impacted smart
eco-cities, catalyzing ongoing improvements and driving solutions to address complex environmental
challenges. This aligns with the visionary concept of smarter eco-cities, an emerging paradigm of ur-
banism characterized by the seamless integration of advanced technologies and environmental strate-
gies. However, there remains a significant gap in thoroughly understanding this new paradigm and the
intricate spectrum of its multifaceted underlying dimensions. To bridge this gap, this study provides a
comprehensive systematic review of the burgeoning landscape of smarter eco-cities and their leading-
edge AI and AIoT solutions for environmental sustainability. To ensure thoroughness, the study em-
ploys a unified evidence synthesis framework integrating aggregative, configurative, and narrative
synthesis approaches. At the core of this study lie these subsequent research inquiries: What are the
foundational underpinnings of emerging smarter eco-cities, and how do they intricately interrelate,
particularly urbanism paradigms, environmental solutions, and data-driven technologies? What are the
key drivers and enablers propelling the materialization of smarter eco-cities? What are the primary AI
and AIoT solutions that can be harnessed in the development of smarter eco-cities? In what ways do AI
and AIoT technologies contribute to fostering environmental sustainability practices, and what potential
benefits and opportunities do they offer for smarter eco-cities? What challenges and barriers arise in the
implementation of AI and AIoT solutions for the development of smarter eco-cities? The findings
significantly deepen and broaden our understanding of both the significant potential of AI and AIoT
technologies to enhance sustainable urban development practices, as well as the formidable nature of
the challenges they pose. Beyond theoretical enrichment, these findings offer invaluable insights and
new perspectives poised to empower policymakers, practitioners, and researchers to advance the inte-
gration of eco-urbanism and AI- and AIoT-driven urbanism. Through an insightful exploration of the
contemporary urban landscape and the identification of successfully applied AI and AIoT solutions,
stakeholders gain the necessary groundwork for making well-informed decisions, implementing effec-
tive strategies, and designing policies that prioritize environmental well-being.
©2023 The Authors. Published by Elsevier B.V. on behalf of Chinese Society for Environmental Sciences,
Harbin Institute of Technology, Chinese Research Academy of Environmental Sciences. This is an open
access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
1. Introduction
The rapid advancement and groundbreaking convergence of
Artificial Intelligence (AI) and the Internet of Things (IoT) has trig-
gered profound transformations across various domains, including
*Corresponding author.
E-mail address: simon.bibri@epfl.ch (S.E. Bibri).
Contents lists available at ScienceDirect
Environmental Science and Ecotechnology
journal homepage: www.journals.elsevier.com/environmental-science-and-
ecotechnology/
https://doi.org/10.1016/j.ese.2023.100330
2666-4984/©2023 The Authors. Published by Elsevier B.V. on behalf of Chinese Society for Environmental Sciences, Harbin Institute of Technology, Chinese Research
Academy of Environmental Sciences. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Environmental Science and Ecotechnology 19 (2024) 100330
environmental sustainability, climate change, and urban develop-
ment. This surge has led to the emergence of smarter eco-cities,
where Artificial Intelligence of Things (AIoT) takes center stage
and garners significant attention. In this context, AIoT is poised to
offer innovative solutions to the mounting environmental chal-
lenges confronting smart cities and, by extension, smart eco-cities.
AIoT holds the potential to unlock avenues for improving resource
efficiency, reducing energy consumption, streamlining waste
management, enhancing transportation management, conserving
biodiversity, and mitigating environmental impacts. These ad-
vancements hold the power to reshape the urban development
landscape in response to the surging wave of urbanization and the
increasing complexity of ecological degradation, transforming cit-
ies into hubs of intelligence, sustainability, and environmental
consciousness.
AI is rapidly reshaping both the technological and urban land-
scapes, serving as a pivotal driving force behind the advancement
of smart cities and smart eco-cities (e.g., Ref. [1e4]). Its potential for
disruption and innovation (e.g., Ref. [5,6]) has positioned it at the
forefront of these developments. This influence profoundly alters
the functioning of urban systems and the intricate interactions,
behaviors, and responses of their subsystems to the surrounding
environment. As a result, urban processes and practices are un-
dergoing significant transformation, marked by their increasing
alignment with data-driven scientific urbanism. The impact of AI
on urban systems and activities has been consistently expanding
[7,8], with its computing capabilities experiencing exponential
growth to accommodate the mounting influx of data from diverse
sources facilitated by IoT. The potential of IoT lies in its capacity to
facilitate data analysis through AI models and algorithms, a synergy
poised to catalyze environmentally sustainable urban
development.
The centralized infrastructure of IoT is grappling with significant
challenges, primarily driven by the extensive strain imposed by the
immense volume of data being generated and processed. Har-
nessing actionable insights from these data necessitates the inte-
gration of AI models and algorithms to effectively manage the data
flow, storage, and processing inherent in the IoT infrastructure. The
emergence of AIoT is underpinned by various factors that set it
apart from traditional IoT. To begin with, AIoT capitalizes on the
synergies between AI and IoT technologies, facilitating more
intelligent and efficient data processing, analysis, and decision-
making (e.g., Ref. [9e11]). Through the integration of AI and Ma-
chine Learning (ML)/Deep Learning (DL) capabilities into IoT de-
vices and systems, AIoT empowers real-time data insights,
predictive analytics, and adaptive responses, thereby optimizing
the overall system performance and efficiency. A key distinction
lies in AIoT's ability to overcome the limitations of IoT in managing
the copious and diverse data generated by the vast network of
interconnected devices. Additionally, AIoT addresses the challenges
associated with transmitting the rapid torrent of data from
distributed sensor network infrastructure [12,13]. By harnessing
the power of AI, AIoT effectively processes and contextualizes
intricate and multifaceted data streams, unlocking the potential for
advanced applications. Most notably in certain domains, AIoT paves
the way for autonomous and intelligent decision-making by IoT
devices, enabling them to learn, adapt, and optimize operations in
response to shifting environmental conditions and user re-
quirements. In essence, the emergence of AIoT introduces a realm
of possibilities for innovation, optimization, and automation across
diverse domains, including environmental sustainability, climate
change, and smart cities (e.g., Ref. [1,13e16]).
Similar to AI, AIoT has become integral to the functioning of
smart cities and, hence, smarter eco-cities. Notably, it has demon-
strated innovative potential in addressing complex environmental
challenges. Recent research has concentrated on the practical ap-
plications of AI and AIoT across various domains of environmental
sustainability and climate change (e.g., Ref. [17 e21]). Toward the
end of 2020 onward, this focus has expanded to encompass smart
cities in terms of their management and planning (e.g.,
Ref. [22e26]). Fundamentally, however, there is a strong inter-
connection between smart cities and eco-cities in that they have
significantly influenced one another over the last decade, particu-
larly in the domains of environmental sustainability and climate
change. Eco-cities have long been associated with these two do-
mains, serving as a well-established paradigm of sustainable ur-
banism (e.g., Ref. [27e33]). However, these two domains have
frequently been addressed separately or, more recently, in
connection with smart cities, particularly within the context of AI
and AIoT, instead of being collectively considered within the
framework of smart eco-cities. This suggests that there has been a
strong tendency to prioritize one approach over another, specif-
ically the second approach has been given a little attention. Espe-
cially, eco-cities and, by extension, smart eco-cities serve as
experimental grounds for innovative solutions and environmental
transitions (e.g., Ref. [27,34e36]). The experimentation within eco-
cities extends beyond climate change to encompass energy tran-
sition, resource conservation, transport efficiency, biodiversity
conservation, and experimental simulation and modeling [37e39].
These strategies and principles, in turn, form the fundamental
driving forces behind the evolution of smart eco-cities.
Amid a world increasingly beset by uncertainties, compounded
by the urgency of the climate crisis and the rapid digital trans-
formation of smart cities and eco-cities, catalyzed by the emer-
gence of AI and AIoT technologies, a compelling trajectory is
emerging. This trajectory envisages the integration of these tech-
nologies' applied solutions into smart eco-cities to deal with the
complexity of environmental degradation and climate disruption
dunder what can be termed as “smarter eco-cities.”This visionary
concept entails the implementation of innovative, forward-looking
strategies to reshape the urban landscape of the future. However, it
is worth acknowledging that while AI and AIoT technologies hold
immense potential yet to unlock, they pose environmental risks
and amplify a spectrum of societal, ethical, legal, and regulatory
challenges.
The body of literature exploring AI and AIoT solutions within the
realm of environmental sustainability, climate change, and smart
cities is rapidly expanding. Nonetheless, to the best of our knowl-
edge, no review study has systematically analyzed and synthesized
the existing corpus of knowledge regarding the interconnections
and synergies between these three domains, coupled with their
intersection with smart eco-cities with respect to AI and AIoT
technologies and solutions. Furthermore, while a few recent review
studies have taken a broader perspective on smart eco-cities, none
have ventured into exploring their emerging technological and
environmental solutions from an integrative perspective. Bridging
these gaps, the present study embarks on a comprehensive sys-
tematic review of the burgeoning landscape of smarter eco-cities
and their leading-edge AI and AIoT solutions for environmental
sustainability. To ensure thoroughness, the study employs a unified
evidence synthesis framework that seamlessly integrates configu-
rative, aggregative, and narrative synthesis approaches. To achieve
the overarching aim, the study focuses on the following specific
objectives:
Describe, illustrate, and make meaningful connections between
the fundamental concepts underpinning emerging smarter eco-
cities.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
2
Present the existing work in the field and explain how the
present study differs from it and why this is a “step forward”and
a new contribution to knowledge.
Analyze, synthesize, interpret, and critically evaluate the exist-
ing knowledge in the areas of environmental sustainability,
climate change, and smart cities to derive comprehensive in-
sights into the role of AI and AIoT solutions in the advancement
of emerging smarter eco-cities.
Categorize AI and AIoT solutions based on these three areas and
further evaluate their impact on advancing environmental sus-
tainability goals.
Capture the dynamic landscape of AI and AIoT solutions for
these three areas by identifying emerging trends, innovations,
and novel approaches within the framework of smarter eco-
cities.
Identify and discuss the key challenges and barriers that arise
when implementing AI and AIoT solutions in the development
of emerging smarter eco-cities.
Identify existing gaps and explore relevant avenues for future
research and areas requiring further investigation.
By pursuing these specific objectives, the study endeavors to
provide a holistic and in-depth understanding of the integration of
AI and AIoT solutions within emerging smarter eco-cities, ulti-
mately contributing to the advancement of environmental sus-
tainability practices in urban contexts. Toward this end, it pursues
the following research questions:
RQ1: What are the foundational underpinnings of emerging
smarter eco-cities, and how do they intricately interrelate,
particularly urbanism paradigms, environmental solutions, and
data-driven technologies?
RQ2: What are the key drivers and enablers propelling the
materialization of smarter eco-cities?
RQ3: What are the primary AI and AIoT solutions that can be
harnessed in the development of smarter eco-cities?
RQ4: In what ways do AI and AIoT technologies contribute to
fostering environmental sustainability practices, and what po-
tential benefits and opportunities do they offer in the realm of
smarter eco-cities?
RQ5: What challenges and barriers arise when implementing AI
and AIoT solutions for the development of smarter eco-cities?
By synthesizing the existing evidence and analyzing the state-
of-the-art research to answer these research questions, the sys-
tematic review contributes to consolidating, enhancing, and
transforming the existing knowledge on smart eco-urbanism by:
Uncovering the dynamjc interplay between urbanism para-
digms, environmental solutions, and data-driven technologies
in emerging smarter eco-cities.
Identifying the driving forces behind the materialization of
smarter eco-cities, namely technological advancements, envi-
ronmental concerns, and policy instruments.
Examining the multifaceted roles of AI and AIoT in environ-
mental sustainability, climate change, and smart cities and how
these technologies can be harnessed in the development of
smarter eco-cities.
Exploring the specific ways in which AI and AIoT technologies
propel sustainable development pracrices and advance envi-
ronmental goals.
Highlighting best practices where AI and AIoT solutions can
significantly contribute to improving the environmental sus-
tainability of smarter eco-cities.
Identifying and evaluating a spectrum of challenges linked to
the implementation of AI and AIoT in smarter eco-cities, aiming
to illuminate potential obstacles and devise strategies to miti-
gate or overcome them.
Unveiling uncharted territories and encouraging the exploration
of AI and AIoT solutions to drive large-scale implementations of
smarter eco-cities, propelling the discourse forward and
fostering innovation in sustainable urban development.
Providing a visionary outlook by elucidating how AI and AIoT
technologies contribute to sustainable urban futures and
exploring the transformative potential of these technologies in
redefining urban infrastructures and systems.
In essence, the study not only presents a comprehensive over-
view of the current landscape of smarter eco-cities but also high-
lights the potential of AI and AIoT technologies in shaping the
future of sustainable urban development, in addition to providing a
roadmap for advancing the discourse on smarter eco-cities and
facilitating interdisciplinary collaborations. Moreover, the applied
unified evidence synthesis approach offers a more holistic and
nuanced understanding of the research topic addressed by
enhancing the thoroughness, depth, and breadth of the systematic
review. The insights derived from the systematic review will not
only inform researchers and practitioners in the field but also guide
policymakers and practitioners in making informed decisions
regarding the adoption and implementation of AI and AIoT tech-
nologies in sustainable urban management and planning. Overall,
by highlighting the solutions, opportunities, benefits, and chal-
lenges in the field of smarter eco-cities, the systematic review will
further facilitate the advancement of research, policy, and practice
in pursuing more sustainable and technologically advanced urban
environments.
This study is structured as follows: Section 2 introduces, de-
scribes, and illustrates the key conceptual strands of the study.
Section 3 addresses the research review related to the study. Sec-
tion 4 describes and illustrates the methodology applied in the
study. Section 5 presents the results of the literature analysis and
synthesis. Section 6 provides a detailed discussion, covering key
challenges, open issues, and limitations. Section 7 identifies rele-
vant gaps and presents recommendations for potential research
directions and areas that require more exploration. This study
concludes, in Section 8, with a summary of key findings and
implications.
2. Conceptual background
Key relevant concepts need to be clarified together with their
integrative and synergistic facets. The value of linking these con-
cepts (Fig. 1) lies in facilitating a better understanding of the
foundational underpinnings of emerging smarter eco-cities in
terms of urbanism paradigms, environmental solutions, and data-
driven technologies.
2.1. Smarter eco-cities and their underlying urbanism paradigms
2.1.1. Smart cities
Smart cities have gained significant attention as a potential so-
lution to address sustainability, resource management, and ur-
banization challenges. Numerous attempts have been made to
define the concept of smart cities. They suggest many definitions
and a plethora of directions to smart city development (e.g.,
Ref. [40e42]). The concept has undergone many changes over the
past two decades. In this regard, it promotes from a technology-
oriented approach, i.e., infrastructures, architectures, platforms,
systems, applications, and models, to a people-oriented approach,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
3
i.e., stakeholders, citizens, knowledge, services, and related data.
Accordingly, it encompasses various dimensions, and there is no
universally accepted definition up till now. However, the working
definition for this study is justified by its alignment with the
research objectives and scope. Accordingly, a smart city denotes an
urban environment that leverages advanced technologies and data-
driven approaches to conserve resources, minimize its environ-
mental impact, and enhance overall ecological well-being. It pri-
oritizes energy efficiency, sustainable transportation, waste
reduction, water conservation, environmental monitoring, and
green infrastructure to create a more eco-friendly and livable
environment while fostering economic growth for all residents.
Smart cities are increasingly emphasizing the role of technological
advancements and scalable data-driven solutions to foster sus-
tainable development practices (e.g., Ref. [43e46]). By integrating
technology with environmental stewardship, smart cities strive to
create greener and healthier living environments.
However, smart cities face several challenges that need to be
addressed to ensure their successful implementation as well as
their integration with other emerging paradigms of urbanism. As
mentioned earlier, one of the primary problems in this regard is the
lack of a standardized definition for smart cities. This lack of clarity
has led to confusion and inconsistency in the planning and
implementation of smart city initiatives. Moreover, the existing
smart city infrastructures are not designed to support the inte-
gration of advanced technologies and data-driven systems. They
need to support the connectivity, data collection, and efficient
management of resources, which involves scalability and interop-
erability. As smart cities grow and more devices and sensors are
connected, their infrastructure must handle the increasing volume
of data and the growing number of users. Integration and seamless
communication between different systems and devices are crucial
for the smooth functioning of their infrastructure. In addition, with
the extensive use of data and connected devices in smart cities,
ensuring the security and privacy of their infrastructure becomes
paramount. These infrastructures must have robust security mea-
sures in place to protect against cyber threats and safeguard the
privacy of citizens' data. Several studies (e.g., Ref. [47,48]) highlight
the importance of addressing data security and privacy issues and
device-level vulnerability in the context of smart cities. Further-
more, coordinating and integrating various domains to create a
cohesive and integrated smart city ecosystem is a complex chal-
lenge. Smart cities should strategically use networked infrastruc-
ture and associated data-driven technologies to produce a smart
economy, smart government, smart mobility, smart environment,
smart living, and smart people. More so, the social, ethical, political,
legal, and regulatory challenges facing smart cities have shown to
be difficult to deal with. To address these challenges, a multidi-
mensional approach focusing on technology, citizens, and in-
stitutions is necessary. This includes developing robust technology
infrastructure, implementing effective governance models, and
actively involving citizens in decision-making processes. Overall,
while smart cities hold great potential for advancing sustainable
urban development, they pose significant challenges that require a
comprehensive approach to build successful and inclusive
ecosystems.
2.1.2. Eco-cities
The concept of eco-cities refers to “an urban environmental
system in which input (of resources) and output (of waste) are
minimized,”[49]. With their ubiquity today, eco-cities vary in the
strategies and solutions they prioritize in response to environ-
mental challenges in different urban contexts. They are widely
diverse in conceptualization, implementation, and development.
Thus, there is no definitive definition of an eco-city but rather a
collection of concepts, ideas, and ambitions [32]. Broadly, eco-cities
are urban areas designed and developed with a strong focus on
environmental sustainability and ecological balance [50]. They aim
to minimize their ecological footprint and promote a harmonious
relationship between humans and nature. They prioritize energy
efficiency, renewable energy sources, waste reduction, green
spaces, sustainable transportation, and resource conservation. They
strive to create a sustainable living environment that supports the
well-being of citizens and the surrounding ecosystems while
fostering social inclusivity and economic prosperity. The ultimate
goal of eco-cities is to create resilient, low-carbon, and environ-
mentally friendly urban spaces that contribute to a more sustain-
able future ([51,52]).
2.1.3. Smart eco-cities
Eco-cities manifest themselves into different models based
mainly on applying the principles of urban ecology or combining
the strategies of sustainable cities and the solutions of smart cities.
For the latter, the most prominent of those models are smart eco-
cities, which integrate IoT and Big Data technologies with envi-
ronmental technologies for achieving urban sustainability (e.g.,
Ref. [35,53e56]). Smart eco-cities refer to urban environments that
integrate advanced technologies, data analytics, and intelligent
systems to enhance sustainability, efficiency, and quality of life
while prioritizing environmental well-being. Accordingly, they
leverage data-driven technologies and solutions to promote the use
of renewable energy, Biomass Combined Heat Power (BCHP), sus-
tainable transportation (walking, cycling, car sharing, biogas cars),
eco-cycle waste management, green infrastructure, urban meta-
bolism, sustainable buildings, smart grids, and sustainable urban
planning strategies to minimize environmental impact, conserve
resources, and foster sustainable and resilient urban environments
(see Refs. [53,57] for illustrative case studies).
Unlike smart cities, smart eco-cities go beyond technology-
driven approaches and emphasize environmental sustainability
and ecological balance as central pillars. They focus on the inte-
gration of sustainable practices and advanced technologies, incor-
porate nature-based solutions into urban planning and design, and
engage communities in environmental stewardship. They aim to
create harmonious urban environments that not only leverage
technology but also prioritize preserving and enhancing natural
resources, biodiversity, and ecosystem services. In essence, they
strive for a more holistic and nature-centric approach to urban
development, promoting long-term sustainability and resilience
and preserving natural resources for future generations to create
environmentally friendly and livable urban communities.
2.1.4. Smarter eco-cities
The concept of smarter eco-cities, as specific to this study, de-
scribes smart eco-cities that integrate AI and AIoT technologies and
solutions with environmental technologies and strategies to
Fig. 1. Smarter eco-cities and their underlying urbanism paradigms, environmental
solutions, and data-driven technologies.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
4
maximize the performance of their sustainable systems and inte-
grate them with smart systems given the clear synergies in their
operation. This integration is intended to produce combined effects
greater than the sum of the separate effects of these systems in
terms of boosting the benefits of environmental sustainability.
Smart city systems include smart grids, smart traffic lights, smart
mobility, smart buildings, smart waste management, and smart
environmental monitoring. Smarter eco-cities represent an evolu-
tion or advancement of smart eco-cities. They prioritize environ-
mental sustainability, employing cutting-edge technologies and
innovative approaches to energy, waste, water, transportation, and
urban planning to create more resilient, sustainable, and techno-
logically advanced urban environments. They also emphasize a
more holistic and comprehensive approach to urban development
by integrating the environmental, social, and economic dimensions
of sustainability. Accordingly, they aim to balance environmental
preservation, social equity, and economic prosperity. They leverage
the advanced technologies and solutions offered by smarter cities,
notably AI and AIoT, to optimize urban systems and address com-
plex challenges in a more integrated and intelligent manner. They
are characterized by das derived based on the synthesized studies:
The innovative potential of AI and AIoT technologies for
sustainability;
The enhanced sustainability outcomes enabled by the applied AI
and AIoT solutions;
The synergies between smart city systems and eco-city systems;
The optimized performance and efficiency of smart eco-city
systems; and
The improved practices of urban management and planning.
In sum, while smart cities focus on technology-driven urban
development, smart eco-cities place a stronger emphasis on
achieving environmental sustainability through IoT and Big Data
technologies. Smarter eco-cities go a step further by incorporating
social and economic dimensions, leveraging emerging AI and IoT
technologies for a more holistic approach to urban development.
2.2. Data-driven technologies
2.2.1. IoT, computing models, and big data
The term “IoT”describes the collective network of physical ob-
jects embedded with sensing, processing, communication, and
actuating technologies and capabilities that enable them to ex-
change data with other devices over the Internet or other networks.
These objects, often called “smart devices,”can range from
everyday items to complex systems like city infrastructure. IoT al-
lows these devices to communicate and interact with each other,
collect data, transfer data, and perform automated tasks, leading to
enhanced efficiency, performance, and sustainability in various
urban domains. More and more IoT devices are connected world-
wide daily and feeding vast amounts of data into analytical systems.
It is estimated that 2.5 quintillion bytes of data are being generated
globally daily, which will rise to 463 exabytes by 2025 [58].
Edge, fog, and cloud computing are three interconnected para-
digms that play a pivotal role in the IoT ecosystem in smart cities
and smart eco-cities in the context of environmental sustainability
[53]. Edge computing involves processing data at or near the data
source, often within the device or a nearby gateway. It aims to
reduce latency and improve real-time responsiveness by executing
computations locally. It is particularly useful for applications that
require quick decision-making and low-latency interactions, such
as autonomous vehicles. Fog computing extends the concept of
edge computing by creating a hierarchical architecture that in-
cludes multiple edge devices and gateways. Fog nodes are
strategically placed in proximity to data sources to perform inter-
mediate processing, data filtering, and preliminary analytics before
transmitting relevant data to the cloud. This approach optimizes
bandwidth usage and enhances system performance, making it
suitable for scenarios involving distributed data sources and
resource-constrained devices. Cloud computing involves using
centralized remote servers to store, manage, and process data. It
offers vast computational resources and storage capabilities, mak-
ing it suitable for complex data analytics, machine learning, and
large-scale processing. Cloud computing allows data to be accessed
and analyzed from anywhere with an internet connection, making
it ideal for applications that require extensive computation and
storage capabilities. These computing paradigms collaborate to
create a holistic IoT ecosystem that efficiently manages data pro-
cessing and analysis across different levels of the network
infrastructure.
Big Data refers to extremely large and complex datasets that
cannot be easily managed, processed, or analyzed using traditional
data processing techniques. They are huge in volume, high in ve-
locity, diverse in variety, exhaustive in scope, fine-grained in res-
olution, and relational, among others. Big Data analytics involves
using advanced techniques and technologies to extract valuable
insights, patterns, and correlations from large and complex data-
sets. This process typically involves data collection, storage, pro-
cessing, analysis, and visualization. In summary, IoT and Big Data
are interconnected concepts revolutionizing how we collect,
analyze, and utilize data. IoT enables the connectivity of smart
devices, allowing them to generate vast amounts of data, while Big
Data provides the means to manage, analyze, and derive mean-
ingful insights from this data. Together, they have the potential to
drive innovation, improve decision-making, and create new op-
portunities in diverse fields, including urban development [59].
2.2.2. AI models and techniques
AI is often described as mimicking human intelligent behavior
by creating computers or machines capable of its simulation. The
working definition for this study describes AI as “any device/system
that perceives its environment and takes actions for its goals”[60].
Broadly, an artificially intelligent machine can learn by acquiring
information on the surrounding environment [61], improving
performance with knowledge from experience, and performing
complex tasks in a way that is similar to how humans solve prob-
lems. The capabilities of AI systems involve data analysis and
learning from external data using Natural Computing (NC) and ML
(e.g., Ref. [62,63]); emulating human cognitive functions using
Computer Vision (CV), Fuzzy Logic (FL), Natural Language Pro-
cessing (NLP) (e.g., Ref. [61]); and dealing with the complexities of
human thinking and emotion (e.g., [64]) using decision support,
strategic planning, sequential actions [65], self-learning, and self-
improvement [66]. Concerning NC, it simulates natural phenom-
ena and utilizes natural material as computational media in com-
puters to optimize ML algorithms [62]. Evolutionary computing
(EC) is also used for continuous optimization and in complex
optimization problems involving many variables. It is extensively
applied to optimize ML models [67]. For example, evolution and
ecology as biological phenomena inspire optimization algorithms
[68]. ML can be based on FL in terms of imitating human reasoning
and cognition using 0 and 1 as extreme cases of truth with various
intermediate degrees, characterizing the space between black-or-
white scenarios, or using fuzzy c-means clustering to separate
each data point into different clusters based on probability score
attribution. CV applies ML to take information from visual data by
recognizing patterns and making meaningful decisions based on
that information. In addition, ML overlaps, intersects, or can be
used as a tool for different AI models, such as CV, FL, and NLP.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
5
For example, NLP enables computers to understand, analyze,
manipulate, and generate human language. NLP plays a significant
role in smart cities by enabling efficient and effective communi-
cation between humans and smart systems. NLP techniques
analyze and understand human language in various forms, allow-
ing for intelligent interactions and decision-making. In smart cities,
NLP can be applied in multiple domains, such as smart planning,
smart governance, smart mobility, and smart services. By
leveraging NLP, smart cities can enhance communication channels,
improve service delivery, and gain valuable insights from citizen
feedback, leading to more responsive and citizen-centric urban
environments. Tyagi and Bhushan [69] explore the potential of NLP
in optimizing Information and Communication Technology (ICT)
processes for building smart cities. The study analyzes the archi-
tecture, background, and scope of NLP and presents its recent ap-
plications in various domains. The authors highlight NLP's role in
advancing smart cities and discuss the open challenges in its
implementation, aiming to emphasize NLP as a key pillar in
building smart cities.
Furthermore, AI involves many techniques that have gained
traction over the past few years as part of AI and AIoT applications
for environmental sustainability, climate change, and smart cities.
These techniques include Artificial Neural Network (ANN), Support
Vector Machine (SVM), Linear Regression (LR), Decision Trees (DT),
Random Forests (RF), Adaptive Neuro-Fuzzy Inference System
(ANFIS), Batch-Normalization (BN), Convolutional Neural Networks
(CNNs), Deep Neural Networks (DNNs), and Genetic Algorithm
(GA). As regards ML, among the supervised learning techniques
used for regression, classification, or both are LR, Generalized
Linear Models (LGM), DT, RF, SVM, ANN, and Bayesian Networks
(BN). As to DL, it is a biological neural network or brain-inspired
type of ML that uses DNN, CNNs, and Recurrent Neural Networks
(RNNs) algorithms. Thus, it emulates the way humans gain certain
types of knowledge by collecting, analyzing, and interpreting large
amounts of data and making decisions in a faster and easier
manner. DL techniques leverage neural networks comprising three
fundamental layers: the input layer, hidden layers, and the output
layer. These layers play a crucial role in acquiring data represen-
tation and establishing connections across multiple levels of
abstraction.
2.2.3. AIoT and its system pillars: A data science cycle perspective
To manage and analyze the dynamic and relational data
generated via IoT increasingly requires powerful computational
and analytical capabilities. This has led to the emergence of AIoT, a
technological framework that optimizes the efficiency of IoT oper-
ations, improves human-machine interactions, advances data
management and analytics models, and enhances decision-making
processes. AIoT involves connecting and combining IoT devices and
sensors with AI models and techniques to enable advanced anal-
ysis, enhanced real-time insights, intelligent decision-making, and
autonomous behavior. It acts through control and interaction to
respond to the dynamic environment, a process where ML/DL has
shown value in enhancing control accuracy and facilitating multi-
modal interactions [13]. The synergy between AI and IoT through
Big Data enables smarter and more efficient applications across
various domains, driving innovation and enabling transformative
solutions. IoT produces Big Data, which in turn requires “AI to
interpret, understand, and make decisions that provide optimal
outcomes”[70] pertaining to a wide variety of practical applica-
tions for urban systems in different context of urbanism [1,71]. In
other words, IoT enables data-driven AI analytics to optimally
accomplish complex tasks and extract useful knowledge in the
form of applied intelligence. The resurgence of AI is driven by the
abundance and potency of Big Data, thanks to enhanced computing
storage capacity and real-time data processing speed.
AIoT enables the utilization of AI to incorporate intelligence and
decision-making capabilities into IoT systems and applications. The
AI/AIoT-driven system consists of five pillars: (1) sensing, (2)
perceiving, (3) learning, (4) visualizing, and (5) acting. This is
illustrated in Fig. 2 from a conceptually generic perspective,
implying that this system can be tailored to various applications
depending on their characteristics, requirements, and objectives.
For example, Zhang and Tao [13] present the progress of AIoT
research from four perspectives: (1) perceiving, (2) learning, (3)
reasoning, and (4) behaving in connection with smart trans-
portation, smart buildings, and smart grids. This entails empow-
ering smart things with human-like cognitive and behavioral
abilities to bring them closer to reality, which is essential to system
operation.
An AI/AIoT-driven system is characterized by the ability to
process raw data to extract useful insights to enable better de-
cisions and/or take action. This involves different interrelated
computational capabilities and processes. Machine perception is
the capability of the system to use input data from sensors (e.g.,
vision, audio, proximity, position, tactile, photoelectric, infrared,
light, and ultrasonic) to deduce different facets of the world, e.g.,
object detection/tracking, action recognition, image classification,
semantic segmentation, language recognition, and pose estimation.
There are different sensory information that provides patterns to
the system for it to generate perceptions. Overall, machine
perception aims to translate these data into meaningful informa-
tion, thereby recognizing and interpreting these data by capturing
the sensory information to relate to the real world. The acquisition
of sensory data from the surrounding environment and their cor-
rect interpretation are key inputs to the learning process. The
common state-of-the-art method for learning is based on ML,
which allows the system to learn from experience without explic-
itly being programmed. Learning starts with collecting and pre-
paring data (e.g., sensory, numbers, human pictures, object images,
records, texts) to be used as training data, building the ML model to
be trained on these data, supplying the data, and letting the system
train itself to find patterns or make predictions. The outcome is an
ML model that can be used with different sets of data and can use
the data for predictive, descriptive, prognostic, and prescriptive
functions. For example, the latter function involves suggesting
what actions to take.
Further, there are three subcategories of ML: (1) supervised
learning (trained with labeled data sets by humans to identify ob-
jects or things), (2) unsupervised/semi-unsupervised learning
(finding patterns in unlabeled data such as behaviors or trends),
and (3) reinforcement learning (trained on trial and error to take
Fig. 2. The five pillars of an AI/AIoT-driven system: 1-sensing in charge of collecting
raw data, 2-perceiving in charge of extracting semantically meaningful information
from raw data, 3-learning in charge of learning to predict patterns, 4-visualizing in
charge of communicating key insights, and 5-acting in charge of taking action to
achieve a certain goal.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
6
the best action based on the right decision such as autonomous
driving or automating routine functions). Ullah et al. [72] illustrate
different supervised, unsupervised, and reinforcement learning
algorithms. These add to transduction learning, multitasking
learning [73], federated learning [74], transfer learning [75], and
few-shot learning [76]. Moreover, to enable the system to learn
from the interpreted data and the performed computations to
generate repeatable outputs and reliable decisions commonly re-
quires using large datasets to train ML models, thereby, the rele-
vance of DL, which can be leveraged to improve the learning of the
system and enable it to adapt to varied situations to enhance its
performance. DL has attracted increased attention and has proven
useful for improving the intelligence of AIoT applications to handle
dynamic and complex environments in the context of unsupervised
and reinforcement learning methods [13].
ML is about leveraging data to improve performance on complex
tasks by building decision-making models (e.g., Ref. [14 ,72,77,78]).
Speaking of tasks, Mitchell [79] conceives of ML as “a computer
program learning from experience ‘E’with respect to some class of
tasks ‘T’and performance measure ‘P,’if its performance at tasks in
‘T’as measured by ‘P,’improves with experience E.’” The work-
horses of the decision-making process are ML models and algo-
rithms, given their role in systematically extracting useful
knowledge from data (patterns, correlations, predictions, forecasts,
etc.) that can support decision-making. For example, by using DL
algorithms, the system can conduct real-time analysis of video
streams, identify objects, and detect events with absolute precision
using CV models for real-time traffic monitoring or analysis of
traffic conditions. The recent development of ML is associated with
its ability to apply complex mathematical calculations to colossal
amounts of data in a repetitive and faster manner. Accordingly, the
system renders the decision-making process more data-driven,
accurate, clearer, and faster thanks to ML. It makes decisions
based on the perceived patterns in the data it receives. Examples of
decisions in this regard may include automating a waste manage-
ment process, enhancing an energy operation, optimizing a plan-
ning function, improving an environmental strategy, adjusting a
policy, and evaluating risk.
As the decision-making process is based on data-driven insights,
data visualization becomes important in terms of conveying com-
plex data in such a way as to make it easier for humans to better
comprehend and react to these data. Data visualization entails us-
ing specialized algorithms to generate graphical representations or
visual displays of data using such elements as charts, infographics,
maps, images, animations, and other metaphors. It enables
decision-makers to gain and pull insights more rapidly by
exploring, monitoring, and interpreting data. Examples of data
visualization include city dashboards, cityScore, city metabolism,
and situation rooms.
Finally, the last process of the system is acting to maximize a
certain goal. To interact with the environment and humans, the
system should be able to reason/make inferences and behave.
Acting is associated with actuation mechanisms, which enable
communication in a smarter eco-city environment and perform
output functions. A wide range of actuators constitutes an integral
part of the sub-systems of smarter eco-cities for operations, func-
tions, and services. Actuation aims to execute actions to optimize
different smart systems, such as power grid, building, transport,
traffic, street lighting, waste management, and water distribution.
The optimization occurs through adding, minimizing, adjusting,
and transferring resources. In this respect, actuation is central to
AIoT applications for monitoring things, controlling things, ranging
things, operating things, repairing things, evaluating things, and
assigning things, to name a few. Functions in smart cities “enable
the actuation mechanisms to be employed directly on the IoT-
enabled smart devices”[80]. For a detailed review of smart city
actuators, the reader might be directed to Ref. [81]. However,
developing more response systems and actuators is required to
improve the engineering applications of AI and AIoT and enable
their implementation in emerging smarter eco-cities concerning
the performance and behavior of their physical systems.
The rationale for adopting a data science type of cycle as to the
pillars of the AI/AIoT system is to emphasize the iterative and data-
driven nature of the system in line with most of the synthesized
studies reporting on the relationship between AIoT environmental
sustainability, climate change, and smart cities. By incorporating
data science principles, we seek to highlight the importance of
leveraging large volumes of heterogeneous data in AIoT applica-
tions. This approach enables the extraction of meaningful insights
and the development of predictive models, leading to enhanced
decision-making processes and improved system performance.
Highlighting the data science cycle provides a holistic AI/AIoT
ecosystem perspective. It emphasizes the continuous feedback
loop, where data are collected, analyzed, and fed back into the
system to optimize its functionality and adaptability. This iterative
process aligns with the dynamic nature of AI and AIoT applications,
where data from various sources continuously flow and contribute
to the intelligence and effectiveness of the overall system.
3. A review of related literature studies
In this section, we present a survey of the existing work con-
ducted in the field of emerging smarter eco-cities and their applied
AI and AIoTsolutions. This survey aims to provide a comprehensive
overview of the current state of research, highlighting the key
findings, contributions, and trends in the field. By examining the
existing literature, we aim to identify gaps and opportunities for
further exploration in developing and implementing smarter eco-
cities. This survey serves as a foundation for our comprehensive
systematic review, enabling us to synthesize and analyze the
findings in a structured and rigorous manner.
It was not until more recently that the literature on AI and AIoT
applications for environmental sustainability and climate change
started to grow and extend across many domains and disciplines.
Several reviews have been performed on AI and AIoT in improving
or advancing the different areas of environmental sustainability
and climate change (Table 1).
3.1. AI for environmental sustainability and climate change
The review studies on energy conservation and renewable en-
ergy contribute to understanding various approaches and strategies
for achieving energy efficiency and promoting renewable energy
sources in different contexts. These studies highlight the impor-
tance of technological advancements, policy frameworks, and
behavioral changes in achieving sustainable energy practices. In the
field of water resources conservation, review studies contribute to
the knowledge on water management practices, including efficient
Table 1
A set of literature review studies on AI solutions for environmental sustainability
and climate change.
Areas References
Conservation and renewable energy [78,82e88]
Water resources conservation [89e93]
Waste management [94e96]
Biodiversity and ecosystem services [97e99]
Sustainable transportation [99e101]
Climate change adaptation and mitigation [18,102,103,104,105]
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
7
water use, water conservation strategies, and the impact of climate
change on water resources. These studies emphasize the need for
integrated water resource management and sustainable water use
practices to address water scarcity and ensure long-term water
sustainability. The research review on waste management dem-
onstrates the significant contributions of AI in waste management,
ranging from enhanced waste sorting to optimized collection
routes, predictive maintenance, and Decision Support Systems
(DSS). These advancements can improve resource efficiency, reduce
environmental impact, and promote sustainable waste manage-
ment practices. The review studies on biodiversity and ecosystem
services provide insights into the importance of biodiversity con-
servation and the role of ecosystem services in sustaining human
well-being. These studies emphasize the need for conservation
measures, habitat restoration, and the integration of ecosystem
services into decision-making processes for sustainable develop-
ment. In the context of sustainable transportation, studies shed
light on various aspects of sustainable transportation, including
electric vehicles, intelligent transportation systems, and multi-
modal transportation options. These studies highlight the potential
of sustainable transportation solutions in reducing CO
2
emissions,
improving air quality, and enhancing urban mobility. Finally, in the
domain of climate change adaptation and mitigation, review
studies contribute to understanding the challenges and opportu-
nities in addressing climate change impacts. These studies explore
adaptation strategies, mitigation measures, and policy frameworks
to reduce GHG emissions and build resilience to climate change.
Collectively, these review studies provide valuable insights into
various aspects of environmental sustainability and climate change
and contribute to the knowledge base in their respective fields.
They highlight the importance of adopting holistic approaches,
integrating multiple disciplines, and considering technological and
policy dimensions to address environmental challenges and pro-
mote sustainable practices.
3.2. AI for smart cities
Some reviews have been conducted on the link between AI, IoT,
and Big Data in smart cities from a more general perspective (e.g.,
Ref. [1,106,107]), broadly addressing different domains beyond
environmental sustainability without providing technical details.
Allam and Dhunny [106] focus mainly on the role of AI in building
smart cities, addressing metabolism, governance, and culture, and
identifying the strengths and weaknesses of AI. The study con-
tributes to understanding the intersection between Big Data, AI,
and smart cities. It explores the potential of utilizing these tech-
nologies in the context of smart cities, highlighting their impact on
various aspects, such as resource optimization, urban planning, and
transportation. It provides insights into the challenges and oppor-
tunities associated with integrating Big Data and AI in smart city
initiatives. Bibri et al. [1] examine the research trends and driving
factors of environmentally sustainable smart cities and their
converging AI, IoT, and Big Data technologies. The authors show
that environmentally sustainable smart cities have experienced
rapid growth during 2016e2022, driven by both the digitalization
and decarbonization agenda and the rapid advancement of data-
driven technologies. The study highlights the importance of
addressing the environmental costs and ethical risks associated
with these technologies. The findings provide insights for scholars,
practitioners, and policymakers developing data-driven technology
solutions and implementing environmental policies for smart cit-
ies. Navarathna and Malagi [107] focus on the role of AI in smart city
analysis. The study examines the application of AI techniques in
analyzing the vast amount of data generated by smart city systems.
It explores how AI can enhance decision-making processes,
optimize resource allocation, and improve the overall efficiency
and sustainability of smart cities. It contributes to understanding
how AI can be leveraged to address smart city development chal-
lenges and complexities.
3.3. AIoT: theoretical foundations and practical applications
Few review studies have been carried out on AIoT as an
emerging technological area. Mukhopadhyay et al. [108] highlight
the importance of sensors in IoT systems and their integration with
AI. The authors emphasize the need for efficient, intelligent, and
connected sensors to make smart decisions and communicate
collaboratively. They also mention the emergence of advanced AI
technologies that enable sensors to detect performance degrada-
tion, identify patterns, and promote innovation. The focus is on
sensors, smart data processing, communication protocols, and AI to
enable the deployment of AI-based sensors for future IoT applica-
tions. Shi et al. [11] focus on the convergence of IoT and AI. The
study compares two AI forms, knowledge-enabled AI and data-
driven AI, highlighting their respective advantages and disadvan-
tages. It surveys recent progress in the integration of AI throughout
the IoT architecture, covering the sensing, network, and application
layers. Zhang and Tao [13] explore the concept of AIoT and its po-
tential to empower IoT. The study presents a comprehensive survey
on AIoT, showcasing how AI techniques, particularly DL, can
enhance IoT speed, intelligence, sustainability, and safety. It dis-
cusses the AIoT architecture within the context of cloud computing,
fog computing, and edge computing. It also highlights promising
applications of AIoT and outlines the challenges and research op-
portunities in this field. Both studies contribute to the under-
standing of the convergence of IoT and AI, with the first study
focusing on the general significance and progress in the integration
and the second study specifically exploring the concept of AIoT and
its implications for IoT. Mastorakis et al. [12] take a broader
perspective on the convergence of AI and IoT, covering various
topics related to AI methods in IoT, including research trends, in-
dustry needs, and practical implementation. Their work balances
theoretical concepts and real-world applications through case
studies and best practices. It serves as a comprehensive resource for
researchers and practitioners interested in the integration of AI and
IoT. Both contributions provide insights into the integration of AI
and IoT technologies and the understanding of AIoT and its appli-
cations in specific contexts and the broader IoT landscape. Together,
these studies provide valuable insights into the advancements,
challenges, and potential applications of AIoT, offering a foundation
for further research and development in this area.
The previous review studies on AI and environmental sustain-
ability, AI and climate change, AI and smart cities, and AIoT have laid
a strong foundation for understanding the opportunities, benefits,
and challenges in creating environmentally sustainable and tech-
nologically advanced urban environments. However, with the rapid
advancement of AI and AIoT technologies, there is a need to explore
the specific intersection of these technologies and existing smart
eco-cities and their underlying multifaceted dimensions. The
emerging field of AI and AIoT presents new possibilities for tackling
the complexity of ecological degradation and the climate crisis in
urban areas. By conducting a comprehensive systematic review of
emerging smarter eco-cities and their leading-edge AI and AIoT so-
lutions, we can bridge the gap between these solutions and the
existing research on environmental sustainability, climate change,
and smart cities within the defining context of smarter eco-cities.
Overall, this review is the first of its kind and seeks to bring new
insights into the flourishing field of smart eco-urbanism and extend
the knowledge of its diverse domains by synthesizing a plethora of
studies from multiple sources and disciplines.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
8
4. Materials and methods
A systematic literature review addressed the three questions
guiding the study and, hence, achieved its specific objectives. It
involves retrieving, mapping, aggregating, configuring, and criti-
cally evaluating studies published to address and discuss the
research topic of smarter eco-cities as an interdisciplinary field
[109]. It allows for the mining of relevant information from the
continuously expanding corpus of publications [110 ]. As illustrated
in Fig. 3, the study consists of nine key stages: (1) research focus
and scope definition, (2) literature search, (3) screening and se-
lection, (4) data extraction, (5) critical evaluation, (6) synthesis and
analysis, (7) interpretation and narration, (8) existing gaps and
areas requiring further investigation, and (9) summary and
manuscript preparation.
Regarding stages 2 and 3, we followed the Preferred Reporting
Items for Systematic Reviews and Meta-analyses (PRISMA)
approach for literature search and selection [111,112 ]. Fig. 4 shows
the four-phase flowchart literature search and selection process
related to this approach. Among the available pool of academic
research databases, SCOPUS was selected given its broad coverage
of 455 high-quality peer-reviewed studies related to the topic on
focus that meets strict standards for rigor. This online platform is
one of the most reliable and trustworthy academic literature sources. To retrieve the scholarly literature, we developed a broad-
based search string covering the different topics of the study and
the associated links. Accordingly, the search string included: “smart
eco-cities,”“smart cities AND internet of things,”“smart cities AND
artificial intelligence OR machine learning OR deep learning,”
“smart cities AND environmental sustainability,”“environmental
sustainability AND artificial intelligence OR machine learning OR
deep learning,”“climate change AND artificial intelligence OR
machine learning,”“artificial intelligence of things AND environ-
mental sustainability”,“artificial intelligence of things AND smart
cities,”“artificial intelligence of things AND climate change,”
“artificial intelligence AND smart eco-cities,”“blockchain AND
environmental sustainability,”and “blockchain AND artificial in-
telligence.”These were used to search against the title, abstract,
and keywords of articles to produce initial insights. We then refined
and narrowed the reading scope, focusing on the documents
providing definitive primary information. Accordingly, titles and
abstracts from these documents were screened to select those
focused on the relationships between smart cities, smart eco-cities,
environmental sustainability, climate change, AI and AIoT tech-
nologies, and IoT and Big Data technologies. After excluding over-
laps, 230 documents remained in the database. These were
checked, and 12 papers were excluded as they did not include in-
formation on the relationships in question. Afterward, we explored
citation tracking or reference chaining techniques to uncover
additional relevant sources. Accordingly, the reference sections of
the remaining papers were checked, and 17 other relevant papers
were added to the final database, which included 235 documents in
total. This was considered reliable when conducting a systematic
review [113].
The reviewed articles were published in prominent journals and
conferences in urban planning, sustainable urban development,
computing, and emerging technologies. Among these outlets were
“Sustainable Cities and Society,”“International Conference on
Smart Sustainable Cities,”“IEEE Transactions on Sustainable
Computing,”“Environmental Modeling and Software,”“Cleaner
Production,”“Environment and Urban Systems,”“Renewable and
Sustainable Energy Reviews,”“Applied Energy,”“Sustainability,”
and “Technological Forecasting and Social Change.”These outlets
showcased the relevance and significance of the research in
advancing the understanding and implementation of AI and AIoT
Fig. 3. Flow diagram outlining the process of conducting the systematic review.
Fig. 4. The PRISMA flowchart for literature search and selection. Adapted from
[111,112 ].
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
9
solutions in the context of smart eco-cities.
The literature search was conducted in late March 2023 and
returned 455 documents covering 2015 to 2023. The starting year
was selected because it marked the approval of the 2030 Agenda for
Sustainable Development by the United Nations General Assembly
as an international policy framework for the 17 SDGs. The full
period, 2015e2023, captures the multifaceted nature of the topic of
smarter eco-cities from the perspective of AI and AIoT technologies
concerning environmental sustainability, climate change, and
smart cities. This is determined by an earlier bibliometric study
conducted by Bibri et al. [1], highlighting several urban trends and
events highly relevant to the current study.
Data extraction and synthesis are crucial steps in conducting a
systematic review. These steps involve extracting relevant infor-
mation from the included studies and synthesizing the data to
identify themes and patterns. Concerning stage 4, we developed a
structured Excel spreadsheet that specifies what information needs
to be extracted from these studies, following a deductive approach
to content analysis. This information included study characteristics,
methodological approaches, technological and sectoral domains, AI
models and techniques applied in environmental sustainability and
climate change, applied AI and AIoT solutions in smart cities, link-
ages between smart cities and smart eco-cities, use cases and ap-
plications, and knowledge gaps, among others. In terms of
methodological approaches, for example, studies were qualitative
based on descriptive analysis and literature review, mixed-
methods, and quantitative approaches based on modeling and
simulation regarding AI and AIoT. Moreover, we evaluated the
quality and relevance of the selected documents through a critical
appraisal process. This involved analyzing the methodologies used
in these documents to assess the strengths and weaknesses of the
research approaches. During data extraction, we carefully read and
analyzed each included study to identify and record the relevant
information.
As regards stage 5 and 6, the primary focus of this study was to
identify, make, establish, and project interconnections between the
different dimensions of smarter eco-cities based on the synthesized
studies. The synthesized analysis involved integrating the extracted
data from multiple studies to identify patterns, trends, similarities,
and differences across studies to generate findings or conclusions.
The synthesis approach applied seamlessly integrated configura-
tive, aggregative, and narrative synthesis as qualitative analytical
approaches. Themes were derived from the researchobjectives and
findings from the reviewed studies. The synthesis was performed
based on these themes using an integrated approach. Specificto
this study, this approach attempted to strike a balance between
theoretical, empirical, and practical issues. Accordingly, it included
evidence from case studies, exploratory studies, observational
studies, experimental studies, theory-building studies, statistical
modeling studies, and review studies. The findings of the synthe-
sized studies were merged based on a set of specified conceptual
and descriptive categories. This process entailed integrating and
fusing information from the multiple studies reporting on the
different dimensions of emerging smarter eco-cities and related
direct and indirect linkages. This categorization evolved as more
precise themes were identified and revised (combined, separated,
refined, or discarded). From the identified categories, themes were
organized to offer new interpretations beyond the synthesized
studies' findings using three different dyet complementary d
approaches to synthesis (Fig. 5).
Configurative synthesis involves interpretation during the syn-
thesis process to identify the big picture and construct the overall
meaning, i.e., thematic synthesis [114 ]. It identifies common
themes across the studies and developing a conceptual framework
to explain the interconnections and variations observed in the
findings. It aims to provide a deeper understanding of the research
evidence by exploring the relationships and context in which the
findings occur. On the other hand, aggregative synthesis involves
summarizing and combining multiple research studies to produce
an overall summary of the findings [115 ], where the interpretation
is performed after the synthesis process to frame the findings, i.e.,
thematic summary. It aims to provide a comprehensive overview
and evaluation of the research evidence. In other words, it adds and
leverages evidence to make statements based on particular con-
ceptual positions [114 ]. Overall, configurative synthesis goes
beyond the aggregation of the research findings and focuses on
understanding the underlying patterns, arrangements, or re-
lationships among the research findings [116 ]. As a form of story-
telling, narrative synthesis involves summarizing, explaining, and
integrating the research findings from individual studies through a
narrative approach. It focuses on providing a descriptive and
interpretive account of the research evidence, often using textual
descriptions [117]. It allows for investigating the similarities and
differences between multiple studies and exploring their re-
lationships [118 ]. It aims to combine diverse perspectives and
findings from multiple studies to generate a coherent narrative
highlighting the key concepts, themes, and implications emerging
from the research. In this respect, it may also involve identifying
common themes or patterns across the studies and providing an
overall narrative of the findings.
5. Results
To present the results, we consolidate all pertinent information
concerning emerging smarter eco-cities as an interdisciplinary
field. This consolidation encompasses a wide spectrum of theo-
retical, empirical, and practical evidence and pertains to the various
dimensions of smarter eco-cities, highlighting their synergies in
producing and enhancing environmental sustainability benefits.
Fig. 6 provides a structured and navigable representation of the
results.
5.1. The relationship between data-driven technologies,
environmental sustainability, smart cities, and smart eco-cities
5.1.1. On the early adoption of IoT and Big Data Technologies in
smart cities in the field of environmental sustainability
To become environmentally smarter and more sustainable, both
smart cities and eco-cities have undergone large-scale digital
transformation enabled by the convergence of IoT, Big Data, and AI
technologies. This has occurred at varying degrees and in different
periods, given the specific focus of these two paradigms of
Fig. 5. A framework for unified evidence synthesis and its characteristics.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
10
urbanism in terms of strategies, solutions, and policies. Accordingly,
in the early 2010s, numerous studies addressed the role of ICT in
tackling the challenges of environmental sustainability in the realm
of smart cities in various domains, notably:
Power grids: to deliver energy and manage its production,
consumption, and distribution to reduce costs and increase the
reliability of energy supply (e.g., Ref. [119 ,120]).
Environmental management: to manage natural resources and
related infrastructure to improve environmental sustainability
(e.g., Ref. [119 ,121,122]).
Transportation management: to optimize transport efficiency
and manage mobility by taking into account traffic conditions
and energy usage (e.g., Ref. [121,123e125]).
Waste management: to collect, recycle, reuse, recover, and
dispose of different types of waste (e.g., Ref. [126,127]).
ICT enables the implementation of smart grids, which use
advanced sensors, communication networks, and data analytics to
optimize energy production, consumption, and distribution. This
helps reduce costs, improve energy efficiency, integrate renewable
energy sources, and enhance the reliability and resilience of the
power grid. Moreover, ICT tools and systems facilitate the moni-
toring, analyzing, and managing of natural resources and related
infrastructure. With sensors, remote sensing technologies, and data
analytics, environmental parameters such as air quality, water
quality, and waste management can be monitored and controlled in
real time, enabling more effective environmental sustainability
practices. Also, ICT applications contribute to optimizing transport
efficiency and managing mobility in smart cities. Intelligent trans-
portation systems, traffic management systems, and real-time data
analysis help monitor traffic conditions, optimize routes, and
improve energy usage. This leads to reduced congestion, better
transportation planning, and reduced energy consumption and
GHG emissions. Furthermore, ICT solutions are utilized to improve
waste management processes. These include systems for waste
collection, recycling, and disposal and technologies for waste
monitoring, sorting, and tracking. By optimizing waste manage-
ment operations and promoting recycling and resource recovery,
ICT enables more sustainable and efficient waste management
practices. ICT has played a vital role in transforming cities into
smart and sustainable environments by enabling key advance-
ments to achieve greater environmental sustainability and resil-
ience in urban areas.
Concurrently, smart cities started to focus on embedding the
nextegeneration of ICT into everyday objects and city structures
and systems as part of the early deployment of IoT (e.g.,
Ref. [128,129]), paving the way for merging digital technologies
with urban infrastructures and coordinating and integrating these
through digital instrumentation and hyper-connectivity. One of the
comprehensive theoretical and empirical studies on smart cities
and IoT and Big Data conducted by Batty et al. [128]defined some
goals concerning the development of a new understanding of
environmental issues and the identification of critical problems
relating to transport, energy, mobility, risks, and hazards. The au-
thors additionally identified some challenges in using manage-
ment, control, and optimization processes to connect smart city
infrastructures to their operational functioning and planning.
During the period 2012e2015, smart cities gained traction as a
model for sustainable urban development, with great potential to
improve environmental sustainability based on IoT and Big Data
Technologies (e.g., Ref. [130e136]). This traction stimulated a
debate on how innovative data-driven IoT technologies could effi-
ciently manage natural resources and mitigate environmental im-
pacts in response to the rapid pace of urbanization and its potential
effects on jeopardizing the sustainability of smart cities. Conse-
quently, IoT and Big Data technologies gained further momentum
in the pursuit of environmental sustainability across the different
domains of smart cities, especially transport, mobility, energy,
waste, pollution control, air quality, and planning (e.g.,
Ref. [137e140]). Subsequently, they became essential to the func-
tioning of smart cities (e.g., Ref. [40,141e145]). This was manifested
in the processes and practices of smart cities becoming “highly
responsive to a form of data-driven urbanism”[146 ]. IoT and Big
Data technologies provide the ability to monitor urban operations,
functions, and structures using advanced forms of decision-making
in urban intelligence functions for design and planning to improve
environmental sustainability.
5.1.2. The influence of the IoT and Big Data Technologies of smart
cities on the materialization of smart eco-cities
It is until around 2014 that smart cities started to have a sig-
nificant impact on eco-cities for environmental sustainability (e.g.,
Ref. [57,147e151) thanks to the adoption of IoT and Big Data tech-
nologies as advanced forms of ICT. [54]) traces the evolution of
smart cities and eco-cities over the past two decades regarding how
their conceptual trajectories have converged under “smart eco-
cities”from the mid-2010s onwards. The author highlights how this
new paradigm of smart, sustainable urbanism is set to leverage the
potential of IoT, Big Data, and digital infrastructures to integrate
urban and green visions and policies. Smart eco-cities became
widespread around 2016/2017 (e.g., Ref. [35,152]) as “a potential
niche where environmental and economic reforms can be tested
and introduced in areas which are both spatially proximate …and
in an international context …through networks of knowledge,
technology and policy transfer and learning”([56], p. 1). Ever since,
IoT and Big Data technologies and their applied solutions have
become instrumental in the functioning of smart eco-cities con-
cerning transport, mobility, energy, waste, pollution control, air
quality, and planning.
To expand on the relationship between smart cities and eco-
cities from an empirical perspective, Bibri and Krogstie [53]
examine and compare the eco-city of Stockholm and the smart city
of Barcelona, focusing on the innovative potential of IoT and Big
Data technologies for advancing the goals of environmental sus-
tainability. The authors show that smart grids, smart meters, smart
Fig. 6. An overview of the main conceptual categories identified and their
relationships.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
11
buildings, smart environmental monitoring, and smart urban
metabolism are the main data-driven IoT solutions adopted to
enhance the performance of data-driven smart cities and smart-
eco-cities under what they term “environmentally data-driven
smart sustainable cities.”They also demonstrate the clear synergy
between the eco-city and smart city solutions as to producing
“combined effects greater than the sum of their separate effects d
concerning energy efficiency and conservation improvement,
environmental pollution reduction, renewable energy adoption,
and real-time feedback on energy and material flows.”Pasichnyi
et al. [153] propose, as part of case study research, a novel data-
driven smart approach to the strategic planning of retrofitting
building energy that allows a holistic city-level analysis and as-
sesses change in total energy demand from large-scale retrofitting.
Similar to Stockholm City [53], the energy transition model of
Barcelona as one of the leading smart cities in Europe, aims to
produce a 100 % certified renewable energy supply plan through
smart energy [154]. Many other eco-cities, mainly from Europe and
China, have adopted IoT and Big Data technologies in the domains
of energy, transport, waste, water, and planning (e.g.,
Refs. [35,54,54e56]). A recent comprehensive state-of-the-art re-
view conducted by Bibri [27] on smart eco-cities reveals that the
newly planned and ongoing eco-city project developments are
increasingly trialing innovative smart technologies to improve
several aspects of environmental sustainability and climate change.
This entails leveraging the advantages of eco-cities and smart cities
and capitalizing on the synergies between their approaches and
solutions, ultimately empowering eco-cities to enhance their
environmental performance. Besides, IoT and Big Data technologies
will fundamentally and irrevocably transform the landscape of eco-
urbanism in terms of how eco-cities will be monitored, understood,
analyzed, managed, planned, and governed.
5.1.3. The influence of policy instruments and government
initiatives on the materialization of smart eco-cities
The materialization of smart eco-cities is strongly influenced by,
in addition to technological advancements and environmental
concerns, policy instruments. Technological advancements provide
the tools and capabilities for these cities to integrate and optimize
various urban systems for sustainability. Environmental concerns
drive the need for these cities to adopt smart and eco-friendly so-
lutions. However, policy instruments play a crucial role in shaping
and guiding the development of smart eco-cities. Policy frame-
works, regulations, and incentives create an enabling environment
for implementing sustainable practices and innovative technolo-
gies and ensure the effective integration of smart solutions into
urban environments. The effective use of policy instruments is
essential in harnessing the full potential of technological ad-
vancements and addressing environmental challenges to realize
the vision of smart eco-cities. Indeed, governance and smart eco-
cities are deeply intertwined in a self-reinforcing relationship. To
put it differently, governance is at the core of smart eco-cities (e.g.,
Refs. [35,53,56,57,155]), and its key function is to make and
implement policy. One of the key roles of urban policy das a set of
plans, laws, rules, regulations, and actions dlies in aligning and
mobilizing the stakeholders involved in the governance of smart
eco-cities. Smart eco-cities require targeted policies to drive prog-
ress in various environmental sustainability areas and effectively
implement innovative data-driven solutions. Overall, policy in-
struments play a crucial role in the materialization of smart eco-
cities by providing a framework for planning, implementing, and
regulating various initiatives. Among the policy instruments that
facilitate the materialization of smart eco-cities [35,55,56,154e156]
are:
Government regulations: The government enacts regulations
that mandate specific sustainability standards and requirements
for smart eco-city development, e.g., setting energy efficiency
targets for buildings, enforcing waste management practices,
and promoting renewable energy integration.
Financial incentives: Governments and local authorities provide
financial incentives to encourage the adoption of smart tech-
nologies and sustainable practices, including tax incentives,
grants, and subsidies for smart eco-city projects that incorporate
smart solutions.
Public-private partnerships: Governments can partner with
technology and energy companies, research institutions, and
industry stakeholders to develop and implement sustainable
solutions. These partnerships leverage expertise, resources, and
funding to implement smart applications successfully.
Open data initiatives: Governments can promote sharing data
collected from various sources, such as sensors and IoT devices,
to foster innovation and facilitate evidence-based decision-
making. These initiatives enable researchers, businesses, and
policymakers to access and analyze information to develop
smart solutions and monitor the performance of smart eco-city
projects.
Standards and certification programs: Establishment of these
programs ensures the interoperability, reliability, and safety of
smart solutions in smart eco-cities. Standards organizations and
certification bodies can define technical requirements, data
privacy guidelines, and cybersecurity protocols to promote trust
and facilitate the adoption of advanced technologies in smart
eco-cities.
These policy instruments create a supportive environment for
the materialization of smart eco-cities. They provide guidance, in-
centives, and regulations that drive sustainable development,
promote technological innovation, and enhance the overall quality
of life for residents. As revealed by Joss and Cowley [304], based on
a comparative case study analysis, policy is found to exercise a
strong shaping role in what sustainable development for cities is
understood to be, which helps explain the considerable differences
in priorities and approaches across countries.
5.2. The rise of AI and AIoT in environmental sustainability, climate
change, and smart cities: solutions, use cases, and applications
This subsection is concerned with the solutions, applications,
and use cases in the realm of AI and AIoT technologies and in-
novations. In this context, a solution represents a comprehensive
approach encompassing software, hardware, processes, and stra-
tegies to solve a particular problem. It is implemented in real-world
use cases, which depict practical scenarios illustrating how data-
driven technologies work in action. On the other hand, an appli-
cation refers to specific software designed to perform tasks, often
utilizing both AI and AIoT models and algorithms. While these
terms can overlap, they offer distinct perspectives: applications
focus on software functionality, solutions provide holistic problem-
solving approaches, and use cases offer practical insights into real-
world implementations, collectively advancing data-driven tech-
nologies and innovations.
The use and application of AI and AIoT in environmental sus-
tainability, climate change, and smart cities have dramatically
increased from 2016 onward. This is seen as a sequel to the wide
adoption of IoT and Big Data technologies and solutions in the
various domains of smart cities for advancing environmental sus-
tainability, as documented by many studies (e.g.,
Ref. [137e140,143,157]). As presented below, the empirical, theo-
retical, and literature research on AI and AIoT solutions,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
12
applications, and use cases involve the different areas of environ-
mental sustainability, climate change, and smart cities. These areas
are organized into two main periods: 2016e2019 and 2020e2023,
which were identified based on an earlier bibliometric study con-
ducted by Bibri et al. [1]. This study explores the key research trends
and driving factors behind the emergence of environmentally
sustainable smart cities and maps their thematic evolution over
time. It demonstrates the rapidly growing trend of this emerging
paradigm of urbanism that markedly escalated during 2016e2022
due to the accelerated digitalization and decarbonization agendas
ddue to COVID-19 and the rapid advancement of data-driven
technologies. Accordingly, the two periods were derived based on
the accelerated digitalization of smart cities prompted by COVID-19
in early 2020 and what this entails in terms of harnessing digital
technologies for addressing environmental issues (e.g.,
Ref. [158e160]), thereby the relevance of subdividing the full
period into two distinct sub-periods. Also, several other studies in
early 2020, as mentioned earlier, emphasized the increasing
recognition of the complexity of environment degradation and
climate change challenges and the growing need for more inno-
vative, advanced, and immediate solutions based on emerging
data-driven technologies to tackle them. As to the starting year of
2016, it has marked the materialization of smart eco-cities (e.g.,
Ref. [27,35,152]), as discussed in Section 4.
5.2.1. The first period: 2016e2019
5.2.1.1. Environmental sustainability
Empirical AI research.The following empirical studies have
significantly contributed to the advancement of emerging smarter
eco-cities by applying AI and ML models and techniques to various
aspects of environmental sustainability.
Water resources conservation: ML techniques, such as ANN,
SVM, FL, ANFIS (an ANN-based on FL), LR, and Key-Nearest
Neighbors (KNN), have been applied to predict stream flow
and examine water quality parameters (e.g., Ref. [161,162]).
Some studies have adopted ML techniques and DSS, such as
ANN, DT, GA, FL, and ANFIS, to analyze water chemistry and
assess water quality (e.g., Ref. [163,164 ]). These models and
techniques have also been applied to hydro-meteorological
forecasting and leak detection [165e167].
Energy conservation and renewable energy: Energy-related
studies have harnessed ML and DSS with techniques such as
ANN, FL, SVM, DT, Evolution Strategies (ES), Evolutionary
Computing (EC), BN (an algorithmic method that makes the
training of DNN faster and more stable), and ANFIS for tasks
including energy operation, production, distribution, mainte-
nance, and planning (e.g., Ref. [92,161,168e170]).
Sustainable transportation: ML has been applied to traffic
forecasting using ANN, DT, and time series models [171,172].
Additionally, transport-related studies have employed ML and
DSS with ANN, DT, NC, FL, SVM, LR, and time series models (e.g.,
Ref. [173 ,174]).
Biodiversity conservation and ecosystem services: ML tech-
niques, including FL, GA, ARIES, and BN (an algorithmic method
used for modeling networks in ecosystems), have been used to
address different aspects of biodiversity conservation and assess
ecosystem services (e.g., Ref. [98,175e179 ]).
The insights gained from the studies conducted on water re-
sources conservation can inform effective water resources man-
agement strategies, aiding in the conservation and sustainable use
of water in existing smart eco-cities. By utilizing AI algorithms,
energy conservation and renewable energy studies have contrib-
uted to energy conservation efforts and the integration of
renewable energy sources within existing smart eco-cities. As to
sustainable transportation, related studies have provided valuable
insights for better transportation planning and management,
facilitating more efficient and sustainable mobility within existing
smart eco-cities. The research on biodiversity conservation and
ecosystem services has contributed to a deeper understanding of
the relationships between biodiversity, ecosystem services, and the
sustainability of existing smart eco-cities. The findings and insights
from these studies can inform evidence-based decision-making
and promote sustainable practices in smart eco-city development.
IoT's role within the AIoT framework.IoT, as a fundamental
component of AIoT, significantly contributes to the above areas of
environmental sustainability by providing a robust technical
framework for data collection, analysis, and decision-making.
Data acquisition: IoT devices equipped with sensors and actu-
ators collect real-time data on environmental parameters. This
extensive data acquisition network forms the foundation for
environmental monitoring.
Data transmission: IoT enables seamless data transmission to
centralized platforms, such as cloud computing, ensuring that
data from distributed sources is readily available for processing
and analysis. Connectivity protocols ensure efficient and secure
data transfer.
Big data handling: IoT generates vast amounts of data and AI
algorithms, including ML and DL, process these data to identify
patterns, anomalies, and trends, providing insights into envi-
ronmental conditions.
Predictive analytics: AI-driven predictive models utilize histor-
ical and real-time IoT-driven data to forecast environmental
changes, which enables proactive decision-making and timely
responses.
Automation and control: IoT's integration with AI allows for
autonomous control of systems and processes. For instance,
smart grids can adjust energy distribution based on real-time
demand and renewable energy availability, promoting energy
efficiency and conservation.
Resource optimization: AI algorithms optimize resource allo-
cation and utilization, ensuring minimal waste and maximum
efficiency in such areas as energy consumption, water usage,
and transportation.
Remote monitoring: IoT-enabled devices can be remotely
controlled and monitored, reducing the need for physical in-
terventions and minimizing human impact on sensitive
ecosystems.
Overall, IoT's technical capabilities within the AIoT framework
enable comprehensive data collection, analysis, and automation,
fostering environmentally sustainable practices and informed
decision-making across various domains.
Theoretical and literature AI research.Similarly, a range of
theoretical and literature review studies have made notable con-
tributions in the context of emerging smarter eco-cities across
various domains:
Water resources conservation: Several studies have focused on
water resources conservation by applying ML and DSS,
employing techniques such as ANN, SVM, FL, GA, and ANFIS. Key
studies in this area include those by Mehr et al. [180], Oyebode
and Stretch [90], Sahoo et al. [181], and Valizadeh et al. [182].
Energy conservation and renewable energy: Efforts to conserve
energy and promote renewable sources have been supported by
ML techniques such as LR, ANN, SVM, NC, and EC. Researchers
and scholars like Akhter et al. [82], Alsadi and Khatib [183], Das
et al. [83], Dawoud, Lin, and Okba [184], Khan et al. [84], Youssef,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
13
El-Telbany, and Zekry [185], and Wang and Srinivasan [186]have
made contributions in this field of study.
Sustainable transportation: Related initiatives have been
advanced by applying ML, CV, and DSS utilizing ANN, DT, NC,
SVM, FL, and time series models. Key studies in this area include
those by Jiang and Zhang [187] and Liyanage et al. [101].
Biodiversity and ecosystem services: Biodiversity-related
studies have explored modeling competition and population
dynamics, often employing ML techniques like cellular autom-
ata [188 ]. Additionally, species conservation efforts have
benefited from ML and DSS involving SVM, ANN, GA, and FL, as
demonstrated by the work of Salcedo-Sanz, Cuadra, and Vermeij
[97]. On the other hand, ecosystem services assessment has
leveraged ML methodologies such as ARIES, as exemplified in
the research by Ochoa and Urbina-Cardona [189 ].
The research on water resource conservation has provided in-
sights into effective water resource management strategies.
Regarding energy conservation and renewable energy, the studies
in this area have contributed to optimizing energy operations,
promoting energy efficiency, and fostering the adoption of
renewable energy technologies. The research on sustainable
transportation has provided insights into traffic management,
transportation planning, and improving the overall efficiency and
sustainability of transportation systems. Concerning the studies in
biodiversity conservation they contribute to our understanding of
biodiversity and the sustainable management of ecosystems.
Overall, these theoretical and literature review studies collectively
contribute to the advancement of smarter eco-cities, and their
outcomes provide insights into the development of evidence-based
approaches for building more sustainable and resilient smart eco-
cities.
5.2.1.2. Climate change: empirical AI research. In empirical AI
research on climate change, notable attention has been directed
toward specific areas, including scenario analysis, marine resources
management, and disaster management and resilience. This focus
has yielded significant contributions to advancing smarter eco-
cities:
Scenario analysis: Studies have employed various AI and ML
models and techniques. Noteworthy works have performed
scenario analysis based on ML, EC, and FL using ANN, BN, SVM,
GA, and neuro-fuzzy (e.g., Ref. [190e192]); analysis across sus-
tainability elements based on ML using ANN (e.g., Ref. [193]);
and CO2 emission [194e196] and natural disaster (e.g.
Ref. [197,198], based on ML and FL using ANN, SVM, EC, and
neuro-fuzzy. Studies on ocean and cryosphere (e.g., Ref. [199])
and atmospheric forecasting (e.g., Ref. [200,201]) have applied
ML using ANN. Furthermore, scenario analysis focuses on
developing and analyzing different scenarios of future climate
conditions based on various parameters, data inputs, and as-
sumptions. It considers a range of potential future outcomes for
climate change, including different levels of GHG emissions,
temperature increases, sea-level rise, and other climate-related
factors. In scenario analysis, various modeling techniques are
used, including Generative Adversarial Networks (GANs), DL
(CNNs and RNNs) and NLP, to create and analyze these scenarios.
The goal is to provide insights into the potential consequences of
different climate pathways and to help inform decision-making
and policy development. Scenario analysis is valuable for un-
derstanding the range of possible future outcomes and their
associated risks and impacts. It allows stakeholders to consider
various “what-if”scenarios and plan accordingly for a range of
potential climate-related challenges.
Disaster management and resilience: The domain of disaster
resilience has witnessed significant AI-driven contributions in
prediction and forecasting and resilient infrastructure and ur-
ban planning. The studies carried out by Cheng and Hoang [305],
Choubin et al. [202], and Ji et al. [203] have advanced the field by
employing different AI models and techniques. Among these are
ML in predictive modeling, pattern recognition, and damage
assessment; NLP in sentiment analysis and information extrac-
tion; CV in image analysis and object recognition; AI-driven
Geographic Information Systems (GIS) in spatial analysis and
mapping and visualization; reinforcement learning in resource
allocation optimization; and DSS in dynamic resource
allocation.
Marine resources management: AI research has played a pivotal
role in marine resources management, encompassing water
pollution monitoring, pollutant tracing in water quality, pollu-
tion reduction and prevention strategies, acidification mitiga-
tion, and habitat and species protection. These endeavors have
harnessed various AI models and techniques, as demonstrated
by the works of Lu et al. [204] and Wang et al. [205], to address
challenges related to the sustainable use and conservation of
marine ecosystems. Some of these models and techniques
include ML, DL (e.g., CNNs and RNNs), GA, and ML-based Species
Distribution Models (SDMs), NLP, time series forecasting,
Autonomous Underwater Vehicles (AUVs) and Remotely Oper-
ated Vehicles (ROVs), and DSS.
This body of research underscores the critical role of AI in
addressing climate change challenges, highlighting its potential to
inform smarter eco-city initiatives and promote sustainable urban
development. Specifically, the research on scenario analysis has
provided insights into climate change impacts, vulnerability, and
adaptation strategies; the interdependencies and interactions be-
tween different aspects of sustainability; understanding and
managing CO2 emissions; and enabling the development of stra-
tegies for reducing carbon footprints. Regarding natural disaster
analysis, the key insights gained pertain to predicting and fore-
casting natural disasters, resilient infrastructure planning, and
environmental hazard detection. Disaster resilience studies have
enhanced disaster resilience, enabling effective preparedness,
response, and recovery measures. The studies on marine resources
management have provided new perspectives on tackling different
relevant challenges and enhancing the sustainable management of
marine resources. Overall, these empirical studies provide valuable
insights and advancements in understanding climate change
challenges in the context of emerging smarter eco-cities.
5.2.1.3. Smart cities. During the first period, AI research on envi-
ronmental sustainability was expanding to include smart cities
based on various AI models, including ML, DL, CV, NLP, and robotics
(e.g., Ref. [107,206e209]). It also involved using ML in empirical
studies (e.g., Ref. [174,210]). Concerning climate change in the ur-
ban context, some studies addressed mitigation concerning urban
planning, mobility, and land use (e.g., Ref. [211,212]). For AI-
powered IoT architecture, research tended to focus mainly on the
theoretical aspects of ML and data analysis (e.g., Ref. [213e215]),
cognitive AI (e.g., Ref. [216,217]), knowledge-based DSS [218], and
knowledge-based AI [174]. Regarding the latter, expert systems are
associated with the capability of AI to solve complex problems, e.g.,
assessment of climate change impacts, and to aid decision-making
by relying on specific knowledge derived from databases [17]. AIoT
has been deployed in multiple ways to help users efficiently
manage energy to reduce cost as well as energy producers optimize
their equipment for better service delivery [219]. It also found
important applications in vehicles and transportation, especially
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
14
self-driving or autonomous vehicles. These are embedded with
several sensing instruments (e.g., cameras, Light Detection and
Ranging (LIDAR), and radar) and thus generate massive amounts of
data [220]. LIDAR is a remote sensing technology that uses laser
light to measure distances and generate precise, three-dimensional
representations of objects as well as landscapes.
5.2.2. The second period: 2020e2023
During the second period, 2020e2023, the empirical, theoret-
ical, and literature research on AI increased and expanded across
environmental sustainability, climate change, and smart cities. This
increase and expansion pertain to energy conservation and
renewable energy, water resources conservation, sustainable
transport and mobility, biodiversity conservation, pollution control,
climate adaptation and mitigation, and disaster management and
resilience. The rapidly growing body of research in environmental
sustainability, climate change, and smart cities has made further
valuable contributions and advancements in the context of
emerging smarter eco-cities, enhancing and extending the knowl-
edge gained for the strategies and solutions for addressing and
overcoming environmental challenges.
5.2.2.1. Environmental sustainability. While the main areas of AI
and AIoT research in environmental sustainability continued to
attract attention, academic interest slightly decreased compared to
the previous period. This is likely attributed to COVID-19 taking
priority in research during the second period, inducing scholars in
smart cities and environmental sustainability to investigate the link
between COVID-19 and CO
2
emissions reduction and air quality
improvement. Worth pointing out is that, during 2020e2021,
COVID-19-related publications received 20 % of all citations, and 98
of the 100 most-cited publications were associated with COVID-19
[1]. There was a shift during this period in academic interest from
focusing on environmental sustainability challenges toward
focusing on the massive deployment and implementation of digital
technologies and applied solutions offered by smart cities. How-
ever, the areas of environmental sustainability that continued to
attract AI research during COVID-19 include:
Energy conservation and air quality efforts encompass pollution
reduction and prevention, pollution monitoring, pollutant
filtering and capture, air quality prediction, and early hazard
warning, as well as the promotion of clean and renewable en-
ergy sources (e.g., Ref. [21,86,87,99,221e224]) and environ-
mental quality control [225].
Sustainable transportation initiatives include the evaluation of
energy consumption in household transportation through ML
models [226], energy planning, transportation connectivity,
urban traffic management, assessment of transport network
capacity, urban traffic surveillance, and optimization of
commuting corridors and jobs-housing balance using tech-
niques, such as GA, EC, ANN, Spatial DNA, and reinforcement
learning (e.g., Ref. [17,99,227e230]).
Clean water security and water resource conservation efforts
involve various aspects such as water quality management,
water supply quantity optimization, water control, water
treatment, and sanitation (e.g., Ref. [93,93,231,232]). Several AI
models and techniques have been applied to address challenges
related to clean water security and water resource conservation,
including ML, DL, ANN, CNNs, RNNs, GA, Particle Swarm Opti-
mization (PSO), NLP, and DSS.
Biodiversity and conservation endeavors revolve around
enhancing and protecting natural capital, preserving ecosystem
health, safeguarding habitats, restoring ecosystems, maintain-
ing forest landscape visual quality, protecting species,
conserving biological diversity, preventing marine pollution,
and ensuring the preservation of marine resources (e.g.,
Ref. [99,233e237,238]). Among the AI models and techniques
being used in biodiversity and conservation efforts are ML, DL,
FL, SVM, CNNs RNNs, GA, NLP, ARIES, AI-driven GIS, and AI-
powered drones.
In the pursuit of creating more intelligent and sustainable urban
environments, a wealth of research has emerged to address the
unique challenges emerging smarter eco-cities face. These studies,
spanning diverse domains, have made further contributions to
shaping the trajectory of these cities. The studies on energy con-
servation and air quality studies significantly enhance their envi-
ronmental performance and efficiency. The research in sustainable
transportation research plays a crucial role in promoting eco-
friendly and efficient urban transport systems. Studies on clean
water security and resource conservation are pivotal for ensuring
sustainable water resource management. The research on biodi-
versity and conservation is instrumental in supporting the devel-
opment of eco-friendly and sustainable environments. These
advancements can reshape the landscape of urban development
models beyond smarter eco-cities in response to the growing wave
of urbanization, making cities more intelligent and environmen-
tally conscious.
5.2.2.2. Climate change. Many more studies were conducted during
the second period, indicating increased scholarly interest in climate
change and its relation to AI and AIoT. The digitalized trans-
formation triggered by COVID-19 significantly contributed to
climate actions [158,159]. Consequently, AI and AIoT technologies
gained strong traction because of the accelerated digitalization
prompted by COVID-19.
Mitigation and adaptation.In the face of escalating environ-
mental challenges, the exploration of innovative technologies has
become paramount to addressing the urgent concerns of climate
change. This narrative dives into the realm of climate change
mitigation and adaptation, shedding light on the pivotal role that AI
and AIoT play in reshaping sustainable urban development in this
regard. Climate change has accelerated the need for proactive
measures, particularly in urban areas where high energy con-
sumption contributes significantly to GHG emissions. Extensive
research conducted during the second period highlights the ur-
gency of addressing climate change as cities contribute significantly
to CO
2
emissions through high energy consumption. It emphasizes
the potential of AI to mitigate climate change by integrating
knowledge, design strategies, and innovative technologies. It
further discusses AI applications in transportation, urban energy,
water use, and waste management, showcasing their impact on
reshaping urban planning and design. Overall, it demonstrates that
implementing AI-driven solutions can improve the sustainability of
future cities and contribute to climate change mitigation efforts.
Ivanova, Ivanova, and Medarov [239] acknowledge the growing
influence of AI across various domains and predict its continued
expansion in the coming decades. The authors emphasize the
prevalence of narrow AI, specialized neural networks designed to
solve specific problems, in technical fields. They focus on applying
narrow AI to investigate the impact of climate change on transport
infrastructure, providing guidelines for data collection and AI
modeling. They highlight the controllability and capabilities of
narrow AI, underscoring their potential for studying climate
change's effects on transportation systems.
Furthermore, in connection with the first period, one of the
areas that received more focus in the theoretical and empirical
research on AI applications for climate change is disaster resilience
and management, including early warning systems, resilience and
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
15
planning, and simulation and prediction (e.g., Ref. [19,240,241]).
However, Leal Filho [18] report ondin a systematic review and
survey questionnaire dall studies conducted during 2020e2022
on the relationship between AI and climate change and its oppor-
tunities for adaptation and mitigation, covering several areas that
have attracted research on AI applications. Among the themes
studied by the authors in relevance to the current study while
expanding on those mentioned in the first period are:
Large-scale urbanization impacts under climate change
scenarios.
Eco-services and tradeoffs model valuation for ecosystem ser-
vices quantification.
Water utilization management using and combining Blockchain
and AI.
AI for disaster response, digital response, and disaster
management.
IoT-Based smart tree management.
AI and ML for wildfire evacuations, wildfire prediction and
prevention, wildfire susceptibility mapping, human-caused
wildfire occurrence, risk-reduction strategies for floods and
droughts, conservation planning under climate-changing pat-
terns, and green-roof irrigation optimization.
ML for flood prediction and protection.
AI for improving resilience and preparedness against flood
events impact.
ANN for drought tolerance determination.
Evolutionary Neural Network (ENN) for forecasting carbon
emissions, energy demand, and wind generation.
ANFIS for modeling climate change impact on wind power
resources.
DL for modeling sub-grid processes in climate models.
ML for water security improvement and water demand fore-
casting in cities.
ML for adaptation policy.
AI for sustainable development.
Some of these themes are linked to smart cities and smart eco-
cities for renewable energy, water, biodiversity, climate, planning,
and policy within the framework of SDG 11: Sustainable Cities and
Communities, SDG 9: Clean and Affordable Energy, and SDG 13:
Climate Action. Moreover, AI can be a critical change agent because
it enables climate change mitigation through carbon neutrality in
energy production, distribution, transportation, buildings, con-
struction, and others [242]. Some studies proposed benchmark
datasets with additional modeling components for better climate
change prediction [243]. For example, Samadi [20] notes that the
convergence of AI and IoT has the potential to accurately predict
floods and accelerate the convergence of AI models and techniques
to advance flood analytics research. The author discusses the
workflow of an AIoT prototype, namely Flood Analytics Information
System (FAIS), which integrates ML, NLP, CNNs, and others.
IoT's role within the AIoT framework.In the context of climate
change mitigation and adaptation, IoT as a core component of AIoT
brings critical technical aspects to the forefront. As to data sensing
and collection, IoT devices equipped with sensors monitor relevant
environmental parameters. This continuous data stream forms the
basis for understanding climate change patterns and trends. In
terms of network connectivity, IoT ensures seamless data trans-
mission and communication between devices and central data re-
positories. Robust communication protocols and networks, such as
5G/6G, facilitate rapid data sharing. Concerning data analysis, AI
algorithms process vast datasets from IoT sensors, identifying
climate change patterns and trends. ML/DL models and algorithms
enable predictive analytics for anticipating changes and their
potential impacts. IoT-connected weather and climate monitoring
stations, coupled with AI-driven forecasting models, enable the
development of early warning systems for extreme weather events
and natural disasters. Moreover, IoT devices track energy con-
sumption in real-time as part of climate change mitigation ap-
proaches. AI algorithms analyze these data to optimize energy use,
identify areas for conservation, and promote the integration of
renewable energy sources to reduce environmental impacts. Also,
IoT-enabled infrastructure, such as smart buildings and resilient
urban planning, enhances adaptation efforts. In this regard, sensors
monitor structural integrity and climate-related risks, facilitating
adaptive responses. Furthermore, IoT sensors in ecosystems track
changes in flora and fauna behavior and health. AI algorithms aid in
assessing the impact of climate change on biodiversity and guiding
conservation efforts. Additionally, IoT-connected satellites and
drones equipped with AI-enabled remote sensing technology pro-
vide critical data for monitoring various environmental changes.
Lastly, regarding real-time decision support: IoT systems provide
real-time climate data and actionable insights to decision-makers,
allowing for adaptive strategies and informed policy develop-
ment. In sum, IoT's technical capabilities within AIoT are instru-
mental in climate change mitigation and adaptation efforts. They
enable data-driven decision-making, resource optimization, and
resilience-building, all crucial components in addressing the chal-
lenges posed by climate change.
5.2.2.3. Smart cities: transportation, energy, waste, environmental
management, and the SDGs. Compared to the first period, AI, AIoT,
and Blockchain technologies have proliferated and expanded,
attracting more research interest, especially with their applications
in smart cities. This also implies that smart cities are embracing AI
and AIoT solutions developed initially for the different areas of
environmental sustainability and climate change as separate fields.
AI and AIoT applications.AI and AIoT applications span many
domains of smart cities in the field of environmental sustainability.
In a systematic literature review on smart cities and AI, Yigitcanlar
et al. [244] found that the key contributions of AI to environmental
sustainability areas include:
optimizing energy production and consumption via domotics
(home automation),
predicting the risks of climate change via ML algorithms and
climate models,
monitoring changes in the natural environment via remote
sensing with autonomous drones, and
operationalizing transport systems via mobility-as-a-service
(MaaS).
AI applications related to the latter theme also include the
management of transport systems of cities in terms of shared
autonomous mobility-on-demand [245], autonomous cities [246],
and autonomous vehicles (e.g., Ref. [247e249]). Concerning the
latter, the deployment of AIoT at the network edge and secure trust
models offer potential solutions for the real-time processing of
sensor data to enable fast response to complex scenarios, such as
obstacle avoidance and velocity adaptation [10]. Moreover, Zhang
and Tao [13] synthesize several studies on the application of AIoT
using DL in smart transportation (e.g., traffic participants, traffic
infrastructures, connected logistics, and in-car driver behavior
monitoring) as well as smart grids (grid fault diagnosis, building
management and optimization, load monitoring and scheduling,
and cyber-attack detection). To solve the load forecasting problem
in energy management, Han et al. [250] propose a DLebased
framework to predict future energy consumption in smart resi-
dential homes and industries, where the IoT network is connected
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
16
to smart grids to maintain energy demand and supply activities
effectively. The experimental results demonstrate the ability of the
approach to predict energy consumption with high accuracy. El
Himer et al. [251] address the role of AIoT in providing new op-
portunities in distributed energy resources (DER), focusing on AIoT
applications in renewable energy sources, such as solar and wind.
An AIoT system developed by Puri et al. [16] generates energy from
different sensors, such as piezoelectric sensors, including from
stress caused by human body weight, heat generated by the
movement of the human body, and sunlight. The authors built and
validated the data collected from the sensors with ANN and ANFIS
models to predict generated power output and demonstrated that
their system could produce accurate results in predicting the power
generated from renewable resources.
Broadly, AIoT can be used in smart cities to analyze and track
how different consumers and residents use energy to make de-
cisions on where and what kind of renewable energy sources could
be used, as well as where energy is being wasted and how it can be
directed for other uses or conserved. Sleem and Elhenawy [252]
discuss the contribution of AIoT to the development of smart
buildings and their functionality, as well as its benefits for reducing
energy consumption and costs, improving occupant comfort and
productivity, and increasing safety and security. The authors also
address the challenges associated with deploying AIoT and
emphasize the potential of AIoT-empowered smart buildings to
contribute to sustainable urban development and improve the
quality of life. Furthermore, AIoT applications are converging in
smart cities. Seng et al. [10] review and discuss several dimensions
of AIoT applications. These are more relevant to this study denergy
and smart grids, industry and smart buildings, vehicles and smart
transportation, and robotics and computer vision.
Furthermore, AI and AIoT have been instrumental in developing
advanced waste collection systems that optimize several parame-
ters and maximize efficiency. Fang et al. [94] provide a compre-
hensive review of the application of AI in waste management,
including waste-to-energy, smart bins, waste-sorting robots, waste
generation models, waste monitoring and tracking, waste logistics,
waste disposal, waste resource recovery, waste process efficiency,
waste cost savings, and improving public health. The authors
highlight the benefits of AI in waste logistics in terms of reducing
transportation distance and time savings, as well as improving
waste pyrolysis, carbon emission estimation, and energy conver-
sion. They also emphasize the role of AI in increasing efficiency and
reducing waste identification and sorting costs in smart cities. Nasir
and Aziz Al-Talib [95] discuss the challenges in waste classification
and the potential of AI and image processing techniques to address
them. They acknowledge the limitations of current waste classifi-
cation models driven by DL and highlight the need for improve-
ments in accuracy and runtime to achieve precise results. They
argue that accurate waste classification is crucial for multiple rea-
sons, including enabling recycling and resource recovery, safe-
guarding the environment and human health, and minimizing
waste management costs. The core idea distilled from the study is
that waste is the byproduct of various human activities, encom-
passing domestic, agricultural, and industrial sectors. Different
types of waste exist, including non-biodegradable, hazardous, in-
dustrial, municipal solid, and agricultural waste. Solid waste can
take hundreds of years to decompose, posing environmental risks.
Mounaded et al. [96] focus on applying AI techniques in municipal
solid waste (MSW) management. They emphasize the use of ANN in
various MSW-related problems and highlight the challenges
related to data reliability and the absence of clear performance
baselines for assessing AI approaches. Overall, these studies
contribute to understanding how AI can revolutionize waste
management by improving waste logistics, classification, and
treatment processes. They highlight the potential benefits and
challenges of implementing AI in the field, providing valuable in-
sights for future research and practical applications.
In particular, the need to overcome the constraints and com-
plexities associated with conventional approaches (e.g., RFID, GPS,
GIS), especially the status and waste level in bins, has driven the
development and implementation of various advanced techniques.
These include PSO (e.g., Ref. [253]), ANN (e.g., Ref. [254]), and
Backtracking Search Algorithm (BSA) (e.g., Ref. [255]) for waste
collection optimization. GA and nearest neighborhood search al-
gorithms have also been used for waste vehicle routing (e.g.,
Ref. [256]). However, these techniques still lack precision and
require a long execution time. Therefore, new techniques are
needed to deal with cost and emission issues and consider bin
capacity, waste weight inside the bin, collection frequency, vehicle
capacity and maintenance, and trip rate [257]. Further, however, AI
models can be applied to predict equipment failures in waste
management facilities. AI models can identify potential issues in
advance by analyzing data patterns, allowing for timely mainte-
nance and minimizing downtime. Also, AI models can enable
powerful DSS for waste management. These systems integrate
various data sources, including weather conditions, population
density, and waste composition, to provide insights and recom-
mendations for effective waste management strategies.
Blockchain and AI and IoT applications.Blockchain technol-
ogy is gaining widespread popularity across various domains,
including energy, environmental conservation, and urban devel-
opment, owing to its capacity to decentralize data and processes
while ensuring robust security measures. In essence, Blockchain is
an open-source, peer-to-peer, distributed ledger system that en-
compasses multiple transactions and their associated data orga-
nized within a chain of interconnected blocks within a
decentralized, peer-to-peer, and openly accessible network, using
technologies such as AI, ML, and Big Data (e.g., Ref. [258]). These
blocks are subject to cryptographic validation by the network itself.
According to Parmentola [259], Blockchain is a rapidly evolving
approach that enables the recording, sharing, updating, and syn-
chronizing of information and transactions across multiple data
ledgers or databases within a distributed and openly accessible
network of diverse participants. Consequently, it fosters enhanced
collaboration and interaction among various organizations and
individuals participating in the network. Moreover, it is distin-
guished by its core attributes, including anonymity, transparency,
auditability, permanence, persistence, and decentralization, which
collectively translate into improved operational performance, effi-
ciency gains, and cost reductions [258,260].
In more recent years, Blockchain has become an innovative
technology and solution for smart cities in environmental sus-
tainability. It has been used by many governments to improve
environmental sustainability. Integrating blockchain into renew-
able energy sources can unlock energy sustainability by facilitating
the development of a decentralized and democratized energy
system while aiding in improved climate governance through its
attributes of transparency, global decentralization, and collabora-
tive capabilities [261]. As demonstrated in a recent bibliometric
study on environmentally sustainable smart cities, Blockchain is
linked to the challenges, services, and resources of smart cities and
AI, IoT, and big data analytics [1]. For the latter, in a use case of a
developed “blockchain-based carbon emission rights verification
system,”AI and Big Data are used to learn proven data [262]. Miao
et al. [263] propose a blockchain and AI-based architecture for the
natural gas intelligent IoT to address the supply chain failure of
existing centralized energy supply architectures because of their
overwhelming numerous requests that could cause pressure,
temperature, and natural gas load to exceed safety limits. Also,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
17
based on multi-sensor-driven (or IoT-based) AI tools, blockchain
platforms can optimize circular economy loops ([264,265]),
allowing carbon footprint reduction and solid waste disposal con-
trol and thereby contributing to sustainability transitions [266].
Xiao et al. [267] propose using Blockchain for intelligent driving
edge systems. The approach utilizes a double auction mechanism to
optimize the satisfaction of users and service providers for edge
computing and shows potential for better performance for resource
utilization.
The value of Blockchain technology lies in storing data on green
energy production activities related to environmental degradation
and air pollution; enabling new means of green energy production,
supply chain and logistics, real-time data collection and analysis for
timely decision-making pertaining to green and low-carbon pro-
cesses; and monitoring EV charging systems [259]. Moreover,
Blockchain informs consumers and users about the use of less-
efficient appliances. It enables them to improve their consump-
tion behavior and thus reduce their carbon emissions [268], in
addition to monitoring compliance with environmental standards
by utilizing product traceability that can decrease resource in-
efficiencies and losses at different supply chain stages [269]. Also,
Blockchain-based initiatives have been designed to provide cred-
ible trading services for polluters. They have also been used to
tokenize carbon credits [269] and monitor carbon emissions. In a
recent review of Blockchain applications in sustainable and smart
cities, Makani et al. [270] provide a detailed account of how
Blockchain technologies contribute to various urban domains,
including transportation, smart grids, smart operational manage-
ment, and smart homes. As regards the trading of renewable energy
by local energy producers based on cryptocurrencies, Blockchain
contributes to energy supply diversification, supply disruption risk
reduction [269], and renewable energy promotion through micro-
grids and other alternative models [259]. These suggest the po-
tential for innovative approaches that support localized and
community-driven renewable energy production. By improving
the alignment of energy supply and demand, these approaches can
strengthen energy security and resilience, offering new avenues to
enhance sustainable energy prospects. Concerning carbon emis-
sions monitoring, the combination of Blockchain and IoT provides
reliability for data and establishes measurement criteria homoge-
neity about registration systems and measurement tools, respec-
tively. These solutions “minimize registration errors and eliminate
fraud arising from the accounting and measurement of gas emis-
sions”[269].
Regarding the role of AI in preventing and reducing marine
pollution [271], Blockchain monitors water pollution changes and
preserves marine resources [259]. It is also used in rewards
schemes for residents of coastal areas using tokens of crypto-
currencies, which “can later be redeemed for credit to collect and
share data on environmental conditions of water bodies”that can
aid in enhancing decisions and designing regulations [269]. Similar
reward schemes can raise public awareness and increase public
participation in waste management and recycling by developing
models to reward active users. Further, implementing Blockchain
and AI as smart city technologies has several co-benefits associated
with water management. Contributions in this regard relate to
water provision efficiency, wastewater management, ensuring
water security, groundwater monitoring, environmental aware-
ness, promoting peer-to-peer trading of water rights, conserving
water resources, and tackling the nexus between various natural
resources in urban areas [259,269,271,272]. In particular, Block-
chain and AI technologies could optimize water management in
water-stressed urban areas by facilitating autonomous water dis-
tribution and management systems These minimize loss and con-
trol quality.
Overall, Blockchain technology, combined with AI and IoT, is
crucial in advancing environmental sustainability. It provides a
decentralized and transparent platform for securely recording and
verifying transactions, data, and information. Integrating AI and IoT
enables efficient data collection, analysis, and decision-making
processes, leading to improved resource management, reduced
environmental impact, and enhanced sustainability practices. The
combination of blockchain, AI, and IoT allows for the development
of innovative solutions such as smart grids, decentralized energy
systems, waste management, and carbon footprint tracking. By
fostering trust, traceability, and accountability, blockchain en-
hances the implementation of sustainable practices and facilitates
the transition toward a greener and more sustainable future.
The sustainable development goals (SDGs).The potential
benefits of smart cities in catalyzing the transition to SDG 11
through advanced technologies and data-driven approaches are
evident. Iris-Panagiota and Egleton [3] explore the role of AI within
smart, sustainable cities, emphasizing its contributions to urban
planning, management, and development. Zaidi et al. [273] analyze
the trajectory of AI in smart sustainable cities research, pinpointing
publication trends and research hotspots, including digital inno-
vation, intelligent data systems, smart energy efficiency, and AI-IoT
data analytics nexus. Yigitcanlar and Cugurullo [26] explore the
sustainability of AI within the context of smart, sustainable cities,
generating insights into emerging urban AI and the potential
symbiosis between AI and smart, sustainable urbanism. The study
reveals that AI applications have become integral in urban services,
managing various aspects of urban life, such as transport systems,
infrastructure, and environmental monitoring. The increasing
adoption of AI is expected to continue, impacting the three di-
mensions of urban sustainability. Vinuesa et al. [271] reveal the
potential of AI to advance 134 targets across all goals while hin-
dering 59 targets. Collectively, these studies enrich the under-
standing of AI's role within smart, sustainable cities, an overarching
umbrella term for smarter eco-cities dits applications in urban
planning, the AI research landscape in smart cities, the integration
of AI and IoT in urban contexts, the alignment of AI with SDG ob-
jectives, and the status of smarter eco-cities. Nonetheless, the
trade-offs of smart cities dprivacy, cybersecurity, digital divide,
technology misuse, and legal frameworks ddemand attention
considering the use of AI and its integration (e.g., Ref. [1,3]). The
imperative lies in devising measures that amplify social and eco-
nomic priorities in smart city planning and development toward
rendering smart eco-cities smarter and more sustainable.
6. Discussion: challenges, open issues, and limitations
6.1. Smart city AI, IoT, and Big Data Technologies as key factors
impacting the dynamics of existing smart eco-cities
AI, IoT, and Big Data technologies are transforming how smart
cities dand hence smart eco-cities dfunction by optimizing their
processes, enhancing their practices, augmenting their solution
capabilities, and improving their environmental sustainability
performance. Since the mid-2010s, the data-driven technologies
and solutions of smart cities as changing elements have gradually
impacted the dynamics of eco-cities toward becoming smart in
their approach to environmental sustainability by integrating their
core domains with smart city domains. This will continue in the
same direction as the AI, IoT, and Big Data technologies and solu-
tions of smart cities become more advanced and integrated with
sustainable technologies and strategies to provide innovative ap-
proaches that can demonstrate the ability to tackle more complex
challenges. This, in turn, means making smart eco-cities smarter in
their pursuit of achieving environmental sustainability thanks to
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
18
the increasing use of AI and AIoT applications. The essence of AIoT
revolves around the need to harness and leverage the power of
smart city technologies and solutions, given the clear synergies in
their operation concerning the optimization, efficiency, manage-
ment, and planning processes of smarter eco-cities. This entails
integrating their systems, coordinating their domains, and coupling
their networks, creating many new opportunities that could be
realized in environmental sustainability.
At the technical level, AI empowers the analysis of the colossal
amounts of data generated via the IoT infrastructure in smart cities
[14] and hence smart eco-cities, largely using ML for decision-
making processes. Regarding AI-enabled sustainable smart cities,
ML models can grow over time, detect invisible anomalies and al-
terations, exhibit various behaviors on different runs for the same
input, and help provide real-time feedback for transport manage-
ment, pollution control, energy management [2], and water man-
agement systems. Intelligent machines can “learn from experience,
adjust to new inputs, and perform human-like tasks”([274], p. 63)
to “interpret external data correctly, to learn from such data, and to
use those learnings to achieve specific goals through flexible
adaptation”([63], p. 17). Overall, AI can provide unsurpassed ways
of automating or autonomizing the repetitive, complex, cognitively
demanding, and time-consuming tasks associated with the oper-
ational functioning and planning of smarter eco-cities.
Concerning environmentally smart sustainable urbanism as an
underlying paradigm of smart eco-cities, it is increasingly
becoming a powerful societal framework for the transition toward
environmental sustainability. This lies in developing joint actions
for preserving the environment based on analyzing large-scale
databases, understanding the complexity of climate change and
modeling and simulating its potential impacts, improving the
health of ecosystems, and enabling high integration of renewable
energy and smart energy [271], enhancing smart renewable energy
infrastructures in smart cities [275], optimizing energy consump-
tion and production, developing more environmentally efficient
transport systems, enhancing environmental monitoring (e.g.,
Ref. [245,276,277]), and strengthening low-carbon energy systems
by supporting circular economies and smart eco-cities [271]. In
particular, more than 250 studies applied AI to energy conservation
and renewable energy during 2015e2019 [17]. The focus on the
potential of AI and AIoT for energy can be justified by its pivotal role
in the transition to smart eco-cities. This occurs through integrating
large shares of renewable energy with smart energy through
additional flexibility and decarbonizing other key emitting sectors,
notably manufacturing, industry, transport, and buildings.
Addressing the energy crisis and reducing fossil fuels will mitigate
the impacts of climate change and make adaptation easier [278].
However, the dynamics of smarter eco-cities should evaluate AI
and AIoT technologies as key components initiating changes in
different domains. Investments in large-scale AI and AIoT as digital
ecosystems are expected to positively impact smarter eco-cities
that may involve feedback mechanisms, resulting in further adop-
tion of these ecosystems and additional future investments. In
other words, AI and AIoT technologies are likely to benefit from
providing innovative applications in response to the need to over-
come environmental sustainability challenges. Accordingly, they
may exhibit positive feedback in that the more their solutions are
implemented, the more likely they will be further implemented,
thanks to network effects, learning, adaptation, and coordination.
While the relationship between outcomes, investments, and
implementations is expected to advance the transition of smart
eco-cities toward environmental sustainability, stating a strong
causal relationship resulting from such linkages needs to be more
accurate. For this reason, coupled with other complex intertwined
internal and external factors, understanding the dynamics of
smarter eco-cities remains a daunting and uncertain challenge. This
can be justified and elucidated in what remains of this discussion.
6.2. Environmental challenges and costs of AI and AIoT technologies
AI, IoT, and Big Data technologies pose significant challenges
when making smart eco-cities environmentally smarter. Therefore,
paying attention to both the opportunities and threats of AI and
AIoT technologies is necessary. Smarter eco-cities must be envi-
ronmentally friendly, thereby minimizing the negative impacts
resulting from the wide use and increasing adoption of the applied
solutions of AI and AIoT technologies. These enabling, integrative,
and constitute technologies are embedded into a much wider
socio-technical landscape involving a complex set of intertwined
and heterogenous factors and actors. There is a risk of a mismatch
between the environmental goals of smarter eco-cities and the
opportunities offered by AI and AIoT technologies. This is due to
their indirect, direct, rebound, and systemic effects, which are
generated through their development, design, use, application, and
disposal (see Ref. [279] for a detailed discussion). In particular, the
indirect effects are expected to be exacerbated the most due to the
increasing demand for AI and AIoT applications. The operation of AI
and AIoT applications requires a lot of energy to power the IoT
infrastructure, data processing platforms, cloud and edge
computing, high-speed wireless networks, and large-scale AI sys-
tems. Concerning the latter, large data centers, which provide
massive computational resources required by AI research, design,
and development, are associated with significant energy con-
sumption and, thus, carbon footprint, compromising the efforts
supporting energy reduction and climate action [271]. According to
current estimates, the global electricity demand for advanced ICT
could increase to 20 % compared to around 1 % today [271]. AI in-
volves establishing heavy energy dependency due to the intensive
use of innovative technologies, increasing environmental impacts
[280,281], extending car traveling distance, and causing urban
sprawl [249]. The latter two relate to exurbanization, a process
whereby upper-class or affluent dwellers move from urban areas to
rural areas to maintain an urban life or live in high-end housing
through advanced technology or long-distance commuting.
However, the high energy requirements for AI and AIoT appli-
cations in the case of the use of non-renewable sources of energy
will undermine the efforts to achieve the environmental targets of
smarter eco-cities. Concerning the direct effects, building smarter
eco-cities requires deploying urban operating systems, urban op-
erations centers, and urban dashboards [53] and thus massive
amounts of natural resources for developing, installing, and
maintaining AI and AIoT ecosystems. In addition, IoT and AI pro-
duction, distribution, service, and disposal produce vast amounts of
e-waste, unsustainable materials, and toxic pollution [63,282].
Almalki et al. [282] discuss, in a recent comprehensive review, the
capabilities and potentials of IoT to respond to the needs of smart
cities while highlighting the challenges for future research on smart
city data-driven IoT applications, with a focus on their risks to
environmental sustainability in terms of energy consumption, toxic
pollution, e-waste, and others. All in all, the applied AI and AIoT
solutions for smarter eco-cities are challenged by the effects of high
energy-intensive structures, undermining the efforts deployed to
avoid the overexploitation of primary resources to achieve carbon
neutrality.
The green growth of AI, IoT, Big Data technologies, green
computing, and eco-friendly design is critical to mitigating the risks
of the mismatch between the environmental goals of smarter eco-
cities and the opportunities offered by AI and AIoT technologies.
This is consistent with the environment being intrinsic to SDG 11 in
terms of recognizing the need to apply the most innovative
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
19
technologies to make critical urban infrastructure resource-
efficient, low-emission, and resilient by reaching the targets
related to energy, climate, transport, waste, and water, as well as
integrated policy and planning. The positive impacts of adopting
sustainable approaches to the development, use, application, and
disposal of AI and AIoT technologies lie in creating eco-friendly
environments that are healthier and more livable in smarter eco-
cities while accelerating their digital transformation. In this
respect, Almalki et al. [282] analyze the various techniques and
strategies for enhancing the quality of life and well-being by
making cities greener, smarter, safer, and more sustainable. Bibri
[157,279] sheds light on the innovative role of advanced ICT as a
potential remedy for mitigating its carbon footprint and thus
advancing environmental sustainability goals, enabling the transi-
tion from smart cities to environmentally smarter cities. In this line
of thinking, Almalki et al. [282] note that the smart things enabled
by IoT in smart cities “become smarter to perform their tasks
autonomously”while communicating “among themselves and
humans with efficient bandwidth utilization, energy efficiency,
mitigation of hazardous emissions, and reducing e-waste to make
the city eco-friendly and sustainable.”Here comes the role of AI in
green computing concerning smarter eco-city sensor integrated
transportation systems, energy systems, building systems, waste
systems, environmental monitoring systems, and so on. Green
computing is key to decreasing carbon emissions and energy con-
sumption to fulfill the environmental goals of sustainability in
smart cities.
It is essential to focus on “reducing pollution hazards, traffic
waste, resource usage, energy consumption, providing public
safety, life quality, and sustaining the environment and cost man-
agement”to make smart cities eco-friendly [282]. This, in turn,
means that AI and AIoT solutions should be carefully implemented
in combination with sustainable and eco-friendly design principles,
energy-efficient policy instruments, and other relevant measures.
This is to ensure that the efficiency gains enabled by AI and AIoT
solutions lead to reducing energy use and carbon footprint. Almalki
et al. [282] provide practical insights into the data-driven IoT-based
eco-friendly and sustainable cities research field. Wang and Liao
[283] explore the intersection of eco-design with AI and Big Data. In
doing so, they identify automation and control systems and com-
puter science among the leading application disciplines. The au-
thors argue for the necessity of more concerted efforts “to advance
both the theoretical and empirical research on the nascent topic
among researchers, funding bodies, policy-makers, and industry
professionals given that the notion of eco-design of AI and Big Data
applications is expected to be pertinent and relevant for designing
greener strategies, products, and services for green digital trans-
formation.”As a justification for a more consolidated green
approach to AI, most attempts at using AI applications to enhance
urban efficiencies have struggled, if not failed, to accomplish the
transformative changes to smart cities due to “the short-sighted,
technologically determined, and reductionist AI approaches being
applied to complex urbanization problems”[4].
6.3. Technical and computational challenges of AI and AIoT
technologies
Inherent to AI models and systems are several technical and
analytical challenges. These include, as indicated by Nishant et al.
[17] in a study conducted on AI for environmental sustainability,
the overreliance of ML models on historical data, the uncertainty
surrounding how humans behave in response to AI-based in-
terventions, and the difficulty in measuring the effects of
intervention strategies. Most new AI systems rely on Big Data and
fail to demonstrate self-ideation or self-creation. AI must develop
new concepts and models of intelligence cognition beyond ML/DL.
This could offer novel solutions for environmental sustainability
and climate change, feeding into new models of smarter eco-cities
that may reduce decision biases due to the incompleteness and
uncertainty of the data collected and aggregated in real-time.
Concerning the over-reliance on Big (historical) Data in smart cit-
ies, Batty et al. (Ref. [128], p. 507) note that the prospect of real-time
data collection and aggregation to deal with urban changes at any
spatial or temporal scale “is a long way off and will never be
reached …but what it does promise is an ability to have a real-time
view of change at different spatial scales and over different time
scales. This will change both the models we can build and how
these technologies can inform the decision process with simula-
tions and decision support being telescoped across space and time.”
While this may be relevant to climate change in modeling and
simulation, human-related variables are, to some extent, unpre-
dictable and dynamically changing. This implies that more and
varied types of data need to be collected and aggregated, new and
more extensive sources of data to be explored, and new and more
advanced tools for handling various velocities of data to be created.
Significantly, historical datasets tend to be of limited value about
climate periods and cycles, which makes it difficult to make precise
predictions or decisions. A deterministic approach is difficult to
adopt in climate change, as it is impossible to estimate ordetermine
potential outcomes precisely. Non-deterministic ML models are
more relevant to transport management, energy management,
water management, waste management, and pollution monitoring,
where they can detect anomalies and alterations and help provide
real-time feedback. Furthermore, the variance-bias tradeoffs asso-
ciated with ML [284] have implications for climate change solutions
due to the bias and oversimplification inherent in predicting future
climate change scenarios.
In addition, Kuguoglu et al. [15] investigate the reasons behind
the failure of many smart city initiatives that rely on AIoT to scale
up. Through a combination of literature study and expert in-
terviews, the study identifies various factors contributing to the
lack of scalability. These factors include resource and capability
constraints, overlooking the importance of comprehensive change,
and the influence of different factors at different stages of imple-
mentation. Regarding the technical challenges related to DL-based
AIoT, they include [13]:
Multimodal heterogeneous data processing, transmission, and
storage pertaining to the massive numbers of heterogeneous
sensors and the vast data streams of different formats, sizes, and
timestamps they generate.
Limited computational and storage resources in relation to using
DL models for real-time data stream processing and low latency.
Computational scheduling in AIoT architecture and related
intense computation. This entails meticulous coordination
across cloud centers, fog nodes, and edge devices, factoring in
variables like data type, volume, network bandwidth, process-
ing latency, performance accuracy, data security, privacy specific
to the application scenario, and unbalanced data flow and user
demands over time.
Labeling unlabeled big data for DL in AIoT in terms of managing
the time-consuming and financially demanding nature of this
process while ensuring high-quality results.
Data monopoly, where access to proprietary data is restricted
due to vested interests, poses challenges to achieving equitable
access to extensive proprietary datasets.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
20
6.4. Challenges and considerations of explainable AI and
interpretable ML
Explainable AI (XAI) and Interpretable ML (IML) encounter sig-
nificant challenges in the context of AI and AIoT solutions for smart
cities, environmental sustainability, and climate change, particu-
larly in the evolving landscape of smarter eco-cities. XAI and IML
are interconnected concepts aiming to enhance the transparency,
comprehensibility, and credibility of AI models for various stake-
holders involved in smarter eco-city development. While XAI fo-
cuses on explaining the decision-making process of AI systems, IML
specifically concentrates on creating ML models that produce easily
interpretable outcomes. XAI encompasses diverse approaches,
including IML techniques, to explain AI decisions, aligning with the
broader aim of enhancing explainability in AI. Both XAI and IML
play pivotal roles in creating AI and AIoT systems that foster
accountability, trustworthiness, and effective human-AI interac-
tion, which is vital for making informed decisions in the context of
smarter eco-cities. Nonetheless, several challenges and issues arise
in this context (e.g., Ref. [285e291,292]), including, but are not
limited to:
Complexity and interpretability: Applying AI and AIoT solutions
to complex challenges in smarter eco-cities can lead to intricate
models, hindering their interpretability. Ensuring these complex
systems generate transparent decisions amid intricate envi-
ronmental and climate data is crucial.
Black-box models: Many advanced AI models (e.g., DNN) are
considered black boxes, lacking transparency in decision-
making. This lack of insight can hinder trust in and adoption
of smart city systems, especially when critical decisions are at
stake.
Bias and fairness: Bias in AI models, derived from biased training
data, can perpetuate existing inequalities in resource allocation
and exacerbate environmental disparities. Overcoming these
biases and ensuring fair outcomes is a daunting task. Biased
decision-making becomes evident in real-time and predictive
analytics, hindering the pursuit of environmental sustainability.
The trade-off between accuracy and interpretability: Striking a
balance between accurate predictions and clear explanations is
essential, as more complex models might offer better pre-
dictions but sacrifice interpretability.
Interdisciplinary nature: Addressing environmental sustain-
ability and climate change requires expertise from diverse fields.
Ensuring that AI and AIoT solutions are interpretable to domain
experts, policymakers, and citizens across various disciplines is
a challenge, as technical terminology can create barriers to
communication.
Data privacy and security: While explaining AI decisions and
promoting transparency, care must be taken not to compromise
sensitive or private information about individuals, potentially
compromising their privacy.
Dynamic and evolving environments: Smart cities and smarter
eco-cities are dynamically changing environments, necessi-
tating adaptable and robust methods for interpreting AI
decisions.
Education and adoption: Educating stakeholders, including
policymakers, city planners, city managers, and citizens, about
the benefits and limitations of AI and AIoT solutions, building
trust and confidence, and encouraging adoption are critical
factors in realizing smarter eco-cities.
The real challenge of XAI lies in granting substantial power to
smarter eco-city systems without simultaneously enabling them to
explain the intricate decision-making processes to different groups
of domain experts. AI and ML models and algorithms assume
control over decision-making by analyzing generated data, pre-
dicting outcomes, and maximizing value based on certain criteria.
This reduces the rich complexity of urban life and the unpredict-
ability of urban dynamics and systems to narrow quantitative and
unitary languages, potentially disregarding the significance of cul-
tural, ethical, social, and political values. As a result, technological
advancements may pose difficulties in achieving the status of
smarter eco-cities due to the mechanistic way of perceiving these
complex systems. Therefore, a recent wave of research has started
to focus on XAI to address some of the concerns posed by the
application of AI in various domains. Mayuri, Vasile, and Indranath
[290] present several applications of XAI/IML and methods to make
AI/ML models explainable/interpretable. Ghonge [287] addressed
several case studies and use cases of XAI as well as its impacts and
challenges in smart city applications. Javid et al. [288] compre-
hensively delve into the landscape of XAI in smart cities, focusing
on current and future developments, trends, enabling factors, use
cases, challenges, and solutions. The authors outline research pro-
jects, standardization efforts, lessons learned, and technical
hurdles.
XAI and IML methods are pivotal for the sustainable advance-
ment of AI and AIoT solutions, allowing society to foster trust in the
environmental and social-economic aspects of sustainability. These
methods explain accuracy, fairness, transparency, accountability,
and human-centeredness outcomes in AI and AIoT-powered deci-
sion-making, addressing ethical and governance concerns. These
principles hold substantial relevance for data-driven decision-
making in smarter eco-cities, thereby the need for creating
explainable/interpretable models, techniques, and tools. Collabo-
rative efforts among AI/AIoT experts, environmental scientists, ur-
ban planners, and policy-makers are essential to ensure the
effective contribution of AI and AIoT technologies to environmental
sustainability and climate change mitigation in the evolving land-
scape of smarter eco-cities. Also, future research endeavors will
play a pivotal role in realizing transparent, effective, and ethically
sound applications of XAI and IML methods within AI and AIoT
solutions, advancing environmental sustainability in smarter eco-
cities while ensuring equitable outcomes for all stakeholders.
6.5. Ethical and Societal Challenges of AI and AIoT technologies
AI technology's ethical and humanistic issues and risks are
subject to long-standing intellectual and philosophical debates. The
development, deployment, and adoption of AI technology raise
these concerns, irrespective of the environmental benefits of its
applied innovative solutions, depending on the application domain.
Against the backdrop of this study, the use of AI involves making
biased decisions, exacerbating privacy and cybersecurity, and
limiting public trust [244,293,294]. Most of these challenges also
apply by extension to AIoT, e.g., AIoT security for smart cities, AIoT
and intrusion detection, and AIoT and trust recommendation [10].
Koffka [9] addresses critical concerns, including security and pri-
vacy, interoperability, and ethics, underscoring the importance of a
responsible AIoT ecosystem. Furthermore, using AI entails devalu-
ating human abilities, deepening information asymmetries,
undermining equal power relations, and causing system failures.
Regarding the latter, increasing public awareness of this type of risk
is crucial before launching large-scale AI deployments in a society
increasingly dependent on AI technology [271]. This indeed is
arcane in that its actual functionalities and mechanisms are un-
derstood by only a group of people, despite being already part of
the everyday life of many of us [6]. Therefore, given that AI as a
disruptive technology will greatly transform smart eco-cities, it
needs to earn public trust regarding how people perceive it. Also, AI
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
21
technology needs to gain the trust in the minds of government
agencies and public organizations investing in it [295]. The chal-
lenges posed by AI generally involve gaps in ethical standards,
including safety, fairness, transparency [293,296], socio-economic
equality, cultural diversity, and social inclusion. For example,
safety is a key topic in ethical and legal debates over autonomous
systems [297,298]. Most ethical issues raised by AI and AIoT ap-
plications relate to the difficulties in explaining AI models or
interpreting ML algorithms, which resemble black boxes, with
some being the hardest for humans to comprehend. There is a need
for developing new methods for Explainable AI (XAI) or Inter-
pretable ML (IML) that allow humans (designers, engineers, re-
searchers, city planners, city managers, regulators, and
policymakers) to understand and trust the decisions or predictions
that AI models and systems make in terms of their potential biases
and expected impacts. Ghonge [287] addressed several case studies
and use cases of XAI as well as its impacts and open challenges in
smart city applications.
The underlying ethical gaps of AI technology call for designing
and implementing appropriate regulatory frameworks to address
the counterproductive outcomes emanating from the penetrative
patterns of AI (e.g., Ref. [63,244,299,300]) in urban life domains in
emerging smarter eco-cities. Especially, early in the decision-
making process of its deployment dwhen the opportunity for
effective inputs and informed choices is greatest. This pertains to
developing “responsible and ethical AI”before it is too late
[297,301,302]. There is a warrant for this as the integration of AI
with IoT and Big Data is speeding up the pace of advancements and
innovations in both AI and AIoT systems, particularly the expo-
nential rise of their computational power, paving the way for them
to gain more and more power of the automation and autonom-
ization of smart cities, with profound implications for smarter eco-
cities. While it is possible to automate certain urban processes and
practices concerning environmental sustainability and climate
change, it is necessary to carefully plan and implement them to
avoid generating fully automated or autonomous smarter eco-cities
based on mechanical decisions. In this regard, it is essential to
address and overcome the regulatory challenges pertaining to the
use of AI and AIoT applications to advance environmental sustain-
ability. Vinuesa et al. [271] emphasize the need for regulatory
insight and oversight to support the development of AI-based
technologies for sustainable development.
The realization of the common good of AI and AIoT technologies
remains highly improbable when AI systems operate solely ac-
cording to the algorithms designed and implemented by powerful
corporations driven by ambitions for power, profit, and extensive
reach and influence. These tech giants pursue various trajectories
and explore uncharted possibilities, raising concerns about the
potential consequences of the large-scale implementation of AI and
AIoT systems. There is an urgent demand for well-regulated and
responsible AI and AIoT systems that prioritize safeguarding public
and civic values within the context of smart eco-cities. Such sys-
tems must be designed to ensure that the broader benefit to society
takes precedence over corporate interests and unchecked ad-
vancements. Indeed, civic values and public values play vital roles
in the functioning of both civil society and government. These
values are the moral compass that guides individuals, communities,
and public institutions in their pursuit of a just, inclusive, and
prosperous society. In a civil society, civic values, such as social
justice, freedom, tolerance, compassion, and tolerance, are the
cornerstones of a harmonious and fair community. They inspire
individuals to engage in civic activities, advocate for their rights,
and work collectively to address societal issues. Civic engagement
is fostered by these values, encouraging citizens to participate in
public discourse and actively contribute to the betterment of
society. At the same time, public values are fundamental to the
proper functioning of government. These principles, including
accountability, transparency, integrity, inclusivity, public partici-
pation, and environmental stewardship, ensure that public in-
stitutions operate in the best interests of the people they serve.
However, for example, in relation to environmental stewardship,
the crucial aspects of environmental protection, justice, and pres-
ervation are often sidelined when unregulated economic interests
drive urban development.
While technological advancements, such as AI and AIoT systems,
have the potential to enhance the efficiency and effectiveness of
both civil society and government, it is essential to recognize that
certain core functions should not be outsourced to these systems.
The decision-making processes guided by civic and public values
require the nuanced judgment and ethical considerations that only
humans can provide. AI and AioT systems can be valuable tools, but
they should support and complement the efforts of individuals and
institutions rather than replace or overshadow the importance of
these foundational values. The interplay between civic and public
values and emerging technologies should be carefully managed to
ensure that they continue to serve as the moral and ethical foun-
dations of our society and government. However, Kassens-Noor and
Hintze [303] argue that the adoption rate of AI technology, coupled
with policy regulations and unforeseen events, has the potential to
transform bustling metropolises into deserted ghost cities. The
complete advancement of AI and AIoT may signify a decline in
moral and societal values, raising concerns about the potential
demise of the human race. Nevertheless, despite the enticing
conceptual and discursive benefits (which relate to both ideas,
theories, and perceptions, as well as the ways in which they are
discussed and communicated) of transitioning cities into eco-cities,
formidable obstacles have hindered large-scale implementations
since the early 1990s, not to mention the development of smart(er)
eco-cities. One of the most significant challenges confronting urban
transformations lies in the significant costs, risks, and uncertainties
associated with the incorporation of AI and AIoT into the realm of
eco-urbanism.
6.6. Methodological limitations
It is important to acknowledge the methodological limitations
of this comprehensive systematic review to allow the readers to
assess the reliability and validity of the findings and understand the
potential implications for future research and practice. These lim-
itations, which arose from the various aspects of the review pro-
cess, include:
Search strategy: Despite efforts to conduct a thorough literature
search, some relevant studies may have been missed. Limita-
tions in database selection, search keywords, language restric-
tion, and inclusion/exclusion criteria could influence the
breadth and depth of the included studies for synthesis. More-
over, as the study relied mainly on peer-reviewed documents,
there is a risk of excluding a large part of grey literature and
stakeholder input and not gaining extensive insights on
emerging smarter eco-cities. This area of research is still
evolving, and most of its existing work drew mainly from the
fields of environmental sustainability, climate change, and
smart cities with respect to AI and AIoT technologies and solu-
tions. Even these fields are associated with a paucity of knowl-
edge and a sparsity of empirical evidence due to the burgeoning
nature of these technologies and solutions.
Publication bias: The study relied on published literature, and
there is a risk of publication bias, where studies with positive
results are more likely to be published, while studies with
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
22
negative findings may be overlooked. This can affect the
comprehensiveness and representativeness of the systematic
review. It is common for researchers and journals to preferen-
tially publish studies based on the direction or significance of
their findings. This can lead to an overrepresentation of studies
with positive results, while studies with negative results may be
less likely to get published and thus be included in the sys-
tematic review. In addition, there is a language bias in that
studies published in English are more accessible and commonly
included in systematic reviews, leading to the potential exclu-
sion of valuable evidence. Especially, English language was one
of the inclusion criteria applied in the study. Funding sources are
another bias regarding the studies funded by industry having a
higher likelihood of being published dif they produce favorable
results.
Data extraction and synthesis pertain to the complexities and
challenges of extracting data from selected studies and syn-
thesizing their findings. One of the primary issues posed in the
study was the heterogeneity across studies. This included vari-
ations in study designs (e.g., methodologies) that complicated
the synthesis of findings and reporting formats (differences in
the presentation of results) that affected the extraction and
synthesis process. These differences can make it difficult to
directly compare and combine data from different studies, thus
obtaining a comprehensive overview of the evidence and
drawing robust conclusions from the systematic review.
7. Suggestions for future research
Smarter eco-city scholars, practitioners, and policymakers have
a new opportunity to foster sustainable development practices
based on a new paradigm of solution-thinking grounded in a
deeper understanding of the interplay between techno-scientific
and socio-political solutions. Developing this multi-faceted
change process is one of the most critical challenges of sustain-
able urban development to achieve the status of smarter eco-cities.
This emerging area of research is empirically under-researched,
theoretically under-developed, and critically under-thought to
allow for large-scale implementations. This means that a plethora
of problems and questions need to be addressed and answered to
guide the development of smarter eco-cities and, hence, large-scale
AI and AIoT deployments for the common good. Some gaps in our
knowledge of emerging smarter eco-cities follow from our results
and discussion. These gaps span a broad set of topics that are sig-
nificant to investigate or critically engage with and that can be
approached from various perspectives in the form of suggestions
for future research.
Considering the transformative potential of AI and AIoT tech-
nologies in reshaping smarter eco-cities, a comprehensive investi-
gation is warranted to unravel the dimensions, opportunities,
benefits, and challenges inherent in this emerging urban paradigm.
Given the nascent stage of research at the intersection of envi-
ronmental sustainability, climate change, and smart cities within
the context of AI and AIoT solutions, the following avenues are
crucial:
Identify key drivers: Delve into the multifaceted drivers that
underpin the evolution of smart eco-cities, encompassing social,
economic, institutional, and political factors beyond techno-
logical and environmental aspects.
Evaluate effectiveness: Prioritize assessing the real-world
effectiveness and scalability of AI and AIoT applications in
smarter eco-cities, determining their actual impact on achieving
SDGs.
Explore long-term benefits: Probe the long-term benefits and
opportunities offered by AI and AIoT technologies in fostering
sustainability practices in smarter eco-cities. This includes
exploring their potential for resource optimization, energy ef-
ficiency, waste reduction, and enhancing the quality of life for
citizens.
Overcome barriers and risks: Confront challenges and mitigate
risks associated with AI and AIoT implementation, such as pri-
vacy concerns, data security, and governance frameworks.
Promote responsible AI practices: Investigate guidelines and
best practices for the responsible design, deployment, and
governance of AI and AIoT solutions. Ensuring fairness, trans-
parency, and accountability in decision-making processes is
essential.
Integrate disciplines: Foster interdisciplinary research merging
environmental sustainability, climate change, and smart cities
with AI and AIoT. This approach unravels intricate relationships
and facilitates a comprehensive understanding of the complex
interactions and interdependencies between these domains,
and enables the development of integrated and holistic
solutions
Formulate policies and frameworks: Develop robust policy and
governance frameworks that facilitate the ethical and trans-
parent use of AI and AIoT technologies and support their
adoption and implementation in sustainable urban develop-
ment. This includes examining regulatory mechanisms, stan-
dards, and guidelines to ensure transparency, accountability,
and ethical use of these technologies.
Advance XAI and IML methods: Develop solutions for inter-
pretability in intricate models, user-centric model training,
tailored solutions for smarter eco-cities, resilience and adapt-
ability enhancement, and ethical implications. This entails
developing XAI techniques for elucidating complex AI models in
AIoT systems, exploring IML integration to engage users in
refining models, utilizing XAI and IML to tackle distinct envi-
ronmental challenges, fortifying AIoT systems' resilience against
dynamic urban scenarios, and scrutinizing the ethical and so-
cietal implications of deploying XAI and IML in AIoT solutions to
ensure equity and transparency.
Promote community engagement: Explore ways to involve local
communities, stakeholders, and citizens in the design, imple-
mentation, and monitoring of AI and AIoT solutions for smarter
eco-cities. Their active participation can lead to more inclusive
and effective outcomes.
Quantify environmental impact: Develop methodologies to
quantify the environmental impact of AI and AIoT solutions in
smarter eco-cities. This involves assessing factors like energy
consumption, carbon footprint, and resource utilization to un-
derstand the overall sustainability gains.
Conduct lifecycle analysis: Assess the sustainability of AI and
AIoT technologies across their entire lifecycle, from production
to disposal. This holistic approach can reveal potential envi-
ronmental hotspots and guide improvements.
Promote citizen participation: Explore ways to empower citi-
zens with AI-augmented information and tools that enable
them to participate in environmental conservation and sus-
tainable behaviors actively.
Enhance behavioral insights: Investigate how AI can leverage
behavioral insights to encourage environmentally friendly be-
haviors among citizens, such as energy conservation and waste
reduction.
Perform case studies: Conduct in-depth empirical inquiries and
practical applications of AI and AIoT in actual smart eco-city
projects. These real-world insights illuminate challenges, best
practices, and valuable lessons for effective integration.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
23
By addressing these knowledge gaps and pursuing these
research avenues, the field of smarter eco-cities can advance its
understanding, implementation, and impact. This will contribute to
developing more sustainable and technologically advanced urban
environments that benefit society and the environment.
8. Conclusion
As disruptive technologies, AI and AIoT lay the foundational
technological infrastructure essential for constructing the digital
ecosystem of emerging smarter eco-cities to amplify and sustain
their contributions to environmental sustainability goals. This
pursuit involves enhancing the efficiency and effectiveness of their
operations, functions, strategies, and policies in alignment with the
environmental targets of SDG 11. Within this context, it is impor-
tant to acknowledge the immense potential of AI and AIoT tech-
nologies to develop robust intelligent systems generating profound
insights for decision-making processes. However, these technolo-
gies cannot serve as a universal remedy or panacea for the wicked
problems characterizing smarter eco-cities as complex systems. In
this study, we aimed to provide a comprehensive systematic review
of emerging smarter eco-cities and their leading-edge AI and AIoT
solutions for environmental sustainability, employing a unified
approach to evidence synthesis. The study's key findings concern-
ing the five research questions are outlined as follows:
Interlinked foundational underpinnings of smarter eco-cities:
The study showed that the fundamental concepts underpinning
smarter eco-cities are intricately interconnected and build one on
another on various scales. The key underlying urbanism paradigms,
namely smart cities and eco-cities, serve as the foundation for
integrating data-driven technologies and environmental solutions.
Data-driven technologies enable real-time monitoring, analysis,
and decision-making, while environmental solutions focus on
optimizing resource efficiency and minimizing ecological footprint.
Data-driven insights enhance the effectiveness of environmental
strategies, ultimately contributing to creating more resilient,
livable, and environmentally friendly urban environments.
The materialization of smarter eco-cities: The study identified
several intertwined factors contributing to the materialization of
smarter eco-cities as an emerging paradigm of urbanism, including
the growing need for sustainable development, advancements in
technology, environmental considerations, policy instruments, and
government initiatives, and the recognition of the potential of data-
driven technologies in addressing complex environmental
challenges.
The primary AI and AIoT solutions harnessed in the develop-
ment of emerging smarter eco-cities: The study identified many
applied solutions of AI and AIoT technologies, demonstrating their
role in urban planning, management, and development. These so-
lutions encompass energy conservation and renewable energy,
sustainable transportation management, traffic control, water re-
sources conservation, waste management for efficient resource
utilization, biodiversity and ecosystem services, environmental
monitoring and control, climate change adaptation and mitigation,
and disaster resilience and management.
The benefits and opportunities of AI and AIoT technologies in
fostering sustainability practices in emerging smarter eco-cities:
The study identified the opportunities and benefits offered by AI
and AIoT technologies in the context of environmental sustain-
ability. Combined, these opportunities and benefits included opti-
mized resource management, increased energy efficiency,
enhanced waste management, improved transportation, and
mobility management, reduced environmental impacts, increased
resilience to environmental challenges, enhanced decision-making
in urban management and planning, and the potential for creating
more sustainable and technologically advanced urban environ-
ments. Identifying opportunities dfavorable circumstances and
possibilities dhelps understand the potential areas where AI and
AIoT solutions can bring about positive changes and contribute to
the overall development of smarter eco-cities. Benefits dpositive
outcomes and advantages dhighlight the tangible and intangible
gains that can be achieved by adopting AI and AIoT solutions.
Challenges and barriers arising in the implementation of AI and
AIoT solutions for the development of emerging smarter eco-cities:
The study identified and evaluated the key challenges pertaining to
environmental costs, privacy concerns related to data collection
and usage, cybersecurity risks related to interconnected systems,
public trust, and social acceptance, limited technical expertise and
knowledge, the lack of robust regulatory frameworks to ensure
ethical and responsible AI and AIoT deployment and the require-
ment for addressing the social issues to ensure equitable and
transparent use of AI and AIoT technologies.
Overall, the study highlighted the significance of AI and AIoT
technologies in advancing the transition toward environmental
sustainability in smarter eco-cities. While these technologies pro-
vide new and largely expanded opportunities to understand better
and prevent environmental problems, they pose significant chal-
lenges that must be addressed and overcome to successfully
implement smarter eco-cities. Therefore, it is important to
emphasize the need for interdisciplinary research, policy support,
and collaboration among different stakeholders to overcome these
challenges and maximize the benefits of these technologies.
The synthesized evidence presented in this study has significant
implications for researchers, practitioners, and policymakers
involved in designing, managing, and planning smarter eco-cities.
It offers valuable insights into the various dimensions of
emerging smarter eco-cities and identifies best practices that can
inform decision-making processes. The systematic review serves as
a knowledge repository, guiding stakeholders in understanding the
current state of research, identifying gaps, and shaping future
strategies for sustainable urban development. Firstly, by identifying
the core conceptual underpinnings of emerging smarter eco-cities
and the intricate interconnections between them (RQ1), coupled
with the intertwined factors propelling the materialization of
smarter eco-cities (RQ2), the study encourages interdisciplinary
collaboration among researchers and practitioners from various
fields and foster new research avenues and practice pathways,
leading to a more thorough understanding and focused improve-
ment of emerging smarter eco-cities. Secondly, by identifying the
key applied solutions of AI and AIoT technologies (RQ3), the study
can inform researchers, practitioners, and policymakers about the
technological advancements and innovative approaches employed
in fostering sustainable urban development practices. Furthermore,
exploring the potential opportunities and benefits offered by AI and
AIoT technologies in this regard (RQ4) can provide valuable insights
for decision-makers and urban planners seeking to leverage these
technologies for achieving the SDGs, especially SDG 11. Lastly,
identifying challenges and barriers in implementing AI and AIoT
solutions in emerging smarter eco-cities (RQ5) can inform policy-
makers and stakeholders about the potential obstacles and open
issues that need to be addressed when integrating these technol-
ogies into urban development strategies. Overall, the research,
practice, and policymaking implications of this study encompass a
wide range of areas, including urban planning, technology imple-
mentation, sustainability practices, and policy development, facil-
itating informed decision-making, and promoting the
advancement of smarter eco-cities.
Ultimately, the findings of the systematic review contribute to
the broader goal of creating smarter eco-cities that prioritize
environmental sustainability, resource efficiency, and human well-
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
24
being. The invaluable insights gained accordingly will empower
stakeholders to make strategic choices, implement innovative so-
lutions, and drive positive change in urban planning and manage-
ment. By leveraging the potential of AI and AIoT, policymakers,
urban planners, researchers, and practitioners can work together
toward creating smarter, more resilient, more livable, and envi-
ronmentally conscious cities that meet the needs of present and
future generations. To sum up, AI and AIoT technologies will offer
unprecedented capabilities to rise to many of the grand environ-
mental challenges, but how these technologies will be used and
what other possible directions this use might take is up to all of us,
especially the research community, and for the time to tell.
CRediT authorship contribution statement
Simon Elias Bibri: Conceptualization, Methodology, Formal
Analysis, Investigation, Data Curation, Visualization, Software,
Writing - Original Draft, Writing - Review &Editing. John Krogstie:
Conceptualization, Writing - Review &Editing. Alexandre Alahi
and Amin Kaboli: Writing - Review &Editing. All authors read and
approved the published version of the manuscript.
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have influ-
enced the work reported in this article.
Acknowledgments
This project has received funding from the European Union's
Horizon 2020 research and innovation program under the Marie
Sklodowska-Curie grant agreement No. 101034260.
Abbreviations
AI Artificial Intelligence
AIoT Artificial Intelligence of Things
ANFIS Adaptive Neuro-Fuzzy Inference System
ANN Artificial Neural Network
ARIES Artificial Intelligence for Ecosystems
AUVs Autonomous Underwater Vehicles
BC Bayesian Classifier
BN Batch-Normalization
CNNs Convolutional Neural Networks
CO
2
Carbon Dioxide
COVID-19 Corona Virus Disease 2019
CV Computer Vision
DL Deep Learning
DNN Deep Neural Networks
DSS Decision Support Systems
DT Decision Trees
EC Evolutionary Computing
ENN Evolutionary Neural Network
ES Evolutionary Strategies
FAIS Flood Analytics Information System
FL Fuzzy Logic
GA Genetic Algorithm
GANs Generative Adversarial Networks
GIS Geographic Information Systems
ICT Information and Communication Technology
IML Interpretable Machine Learning
IoT Internet of Things
KNN K-Nearest Neighbour
LIDAR Light Detection and Ranging
Maas Mobility-as-a-service
ML Machine Learning
NC Natural Computing
NLP Natural Language Processing
RNNs Recurrent Neural Networks
ROVs Remotely Operated Vehicles
SDGs Sustainable Development Goals
SDMs Species Distribution Models
SVM Support Vector Machine
XAI Explainable Artificial Intelligence
References
[1] S.E. Bibri, A. Alahi, A. Sharifi, J. Krogstie, Environmentally sustainable smart
cities and their converging AI, IoT, and big data technologies and solutions:
an integrated approach to an extensive literature review, Energy Inform 6
(2023) 9, https://doi.org/10.1186/s42162-023-00259-2, 2023.
[2] M.K. Gourisaria, G. Jee, G.M. Harshvardhan, D. Konar, P.K. Singh, Artificially
intelligent and sustainable smart cities, in: P.K. Singh, M. Paprzycki,
M. Essaaidi, S. Rahimi (Eds.), Sustainable Smart Cities. Studies in Computa-
tional Intelligence, vol. 942, Springer, Cham, 2023, https://doi.org/10.1007/
978-3-031-08815-5_14.
[3] E. Iris-Panagiota, T.E. Egleton, Artificial intelligence for sustainable smart
cities, in: Brojo Kishore Mishra (Ed.), Handbook of Research on Applications
of AI, Digital Twin, and Internet of Things for Sustainable Development, IGI
Global, 2023, pp. 1e11, https://doi.org/10.4018/978-1-6684-6821-0.ch001.
[4] T. Yigitcanlar, R. Mehmood, J.M. Corchado, Green artificial intelligence: to-
ward an efficient, Sustain. Equit. Technol. Smart Cities Futures 13 (16) (2021)
8952. https://www.mdpi.com/2071-1050/13/16/8952.
[5] A. Ab-Rahman, U.Z. Hamid, T.A. Chin, Emerging technologies with disruptive
effects: a review, Perintis e-J. 7 (2017) 111e128.
[6] A. Greenfield, Radical Technologies: the Design of Everyday Life, Verso Books,
2018.
[7] H. Lu, Y. Li, M. Chen, H. Kim, S.J.M.N. Serikawa, Applications, Brain intelli-
gence: go beyond artificial intelligence, Mobile Network. Appl. 23 (2) (2018)
368e375.
[8] M. Batty, Artificial intelligence and smart cities, Environ. Plan. B Urban Anal.
City Sci. 45 (1) (2018) 3e6, https://doi.org/10.1177/2399808317751169.
[9] K. Koffka, Intelligent Things: Exploring AIoT Technologies and Applications,
Kindle Edition, 2023.
[10] K.P. Seng, L.M. Ang, E. Ngharamike, Artificial intelligence Internet of Things: a
new paradigm of distributed sensor networks, Int. J. Distributed Sens. Netw.
18 (3) (2022), https://doi.org/10.1177/15501477211062835.
[11] F. Shi, H. Ning, W. Huangfu, et al., Recent progress on the convergence of the
Internet of Things and artificial intelligence, IEEE Network 34 (5) (2020)
8e15.
[12] G. Mastorakis, C.X. Mavromoustakis, J.M. Batalla, et al. (Eds.), Convergence of
Artificial Intelligence and the Internet of Things, Springer, Berlin, 2020.
[13] J. Zhang, D. Tao, Empowering things with intelligence: a survey of the
progress, challenges, and opportunities in artificial intelligence of things,
IEEE Internet Things J. 8 (10) (2021) 7789e7817, https://doi.org/10.1109/
JIOT.2020.3039359, 15 May15.
[14] D. Alahakoon, R. Nawaratne, Y. Xu, D. De Silva, U. Sivarajah, B. Gupta, Self-
building artificial intelligence and machine learning to empower big data
analytics in smart cities, Inf. Syst. Front (2020) 1e20.
[15] B.K. Kuguoglu, H. van der Voort, M. Janssen, The giant leap for smart cities:
scaling up smart city artificial intelligence of things (AIoT) initiatives, Sus-
tainability 13 (21) (2021) 12295, https://doi.org/10.3390/su132112295.
[16] V. Puri, S. Jha, R. Kumar, et al., A hybrid artificial intelligence and Internet of
Things model for generation of renewable resource of energy, IEEE Access 7
(2019) 111181e111191.
[17] R. Nishant, M. Kennedy, J. Corbett, Artificial intelligence for sustainability:
challenges, opportunities, and a research agenda, Int. J. Inf. Manag. 53 (2020)
102104.
[18] W. Leal Filho, T. Wall, S.A.R. Mucova, G.J. Nagy, A.-L. Balogun, J.M. Luetz,
A.W. Ng, M. Kovaleva, F.M.S. Azam, F. Alves, Deploying artificial intelligence
for climate change adaptation, Technol. Forecast. Soc. Change 180 (2022)
121662.
[19] M. Raza, M. Awais, K. Ali, N. Aslam, V.V. Paranthaman, M. Imran, F. Ali,
Establishing effective communications in disaster affected areas and artificial
intelligence based detection using social media platform, Future Generat.
Comput. Syst. 112 (2020) 1057e1069.
[20] S. Samadi, The convergence of AI, IoT, and big data for advancing flood an-
alytics research, Front. Water 4 (2022) 786040, https://doi.org/10.3389/
frwa.2022.786040.
[21] W. Sun, P. Bocchini, B.D. Davison, Applications of artificial intelligence for
disaster management, Nat. Hazards 103 (3) (2020) 2631e2689, https://
doi.org/10.1007/s11069-020-04124-3$.
[22] D. Kamrowska-Załuska, Impact of AI-based tools and urban big data analytics
on the design and planning of cities, Land 10 (2021) 1209, https://doi.org/
10.3390/land10111209, 2021.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
25
[23] S.C. Koumetio Tekouabou, E.B. Diop, R. Azmi, et al., Artificial intelligence
based methods for smart and sustainable urban planning: a systematic
survey, Arch. Comput. Methods Eng. 30 (2023) 1421e1438, https://doi.org/
10.1007/s11831-022-09844-2.
[24] T.W. Sanchez, Planning on the verge of AI, or AI on the verge of planning,
Urban Sci. 7 (2023) 70, https://doi.org/10.3390/urbansci7030070.
[25] T.H. Son, Z. Weedon, T. Yigitcanlar, T. Sanchez, J.M. Corchado, R. Mehmood,
Algorithmic urban planning for smart and sustainable development: sys-
tematic review of the literature, Sustain. Cities Soc. 94 (2023) 104562.
[26] T. Yigitcanlar, F. Cugurullo, The sustainability of artificial intelligence: an
urbanistic viewpoint from the lens of smart and sustainable cities, Sustain-
ability 12 (20) (2020) 8548, https://doi.org/10.3390/su12208548.
[27] S.E. Bibri, Data-driven smart eco-cities and sustainable integrated districts: a
best-evidence synthesis approach to an extensive literature review, Eur. J.
For. Res. 9 (1) (2021) 16, https://doi.org/10.1186/s40309-021-00181-4.
[28] Z. Cai, Y. Tang, Toward a sustainable city: a scoping review of eco-cities
development and practices in China, in: L. Bian, Y. Tang, Z. Shen (Eds.),
Chinese Urban Planning and Construction. Strategies for Sustainability,
Springer, Cham, 2021, https://doi.org/10.1007/978-3-030-65562-4_9.
[29] J.R. Kenworthy, The eco-city: ten key transport and planning dimensions for
sustainable city development, Environ. Urbanization 18 (1) (2006) 67e85,
https://doi.org/10.1177/0956247806063947.
[30] M. Mostafavi, G. Doherty, Eco-urbanism, Lars Muller, Baden, Switzerland,
2010.
[31] E. Rapoport, Utopian visions and real estate dreams: the ecoecity past,
present and future, Geogr Compass 8 (2014) 137e149.
[32] Mark Roseland, Dimensions of the eco-city, Cities 14 (4) (1997) 197e202.
[33] R. Register, Ecocities: Rebuilding Cities in Balance with Nature, New Society
Publishers, 2006.
[34] H. Bulkeley, V. Cast
an Broto, Government by experiment? Global cities and
the governing of climate change, Trans. Inst. Br. Geogr. 38 (3) (2013)
361e375.
[35] F. Caprotti, R. Cowley, I. Bailey, S. Joss, F. Sengers, R. Raven, P. Spaeth,
E. Jolivet, M. Tan-Mullins, A. Cheshmehzangi, Smart Eco-City Development in
Europe and China: Policy Directions, University of Exter (SMART-ECO Proj-
ect, 2017.
[36] J. Evans, A. Karvonen, R. Raven, The Experimental City, Routledge, 2016.
[37] K.C. Clarke, Why simulate cities? Geojournal 79 (2013) 129e136, https://
doi.org/10.1007/s10708-013-9499-5.
[38] J.R. Kenworthy, Urban transport and eco-urbanism: a global comparative
study of cities with a special focus on five larger Swedish urban regions,
Urban Sci. 3 (1) (2019) 25, https://doi.org/10.3390/urbansci3010025.
[39] Jonas Torrens, Philip Johnstone, Schot Johan, Unpacking the formation of
favourable environments for urban experimentation: the case of the Bristol
Energy Scene, Sustainability 10 (3) (2018) 1e28, https://doi.org/10.3390/
su10030879.
[40] S.E. Bibri, On the sustainability and unsustainability of smart and smarter
urbanism and related big data technology, analytics, and application, in: Big
Data Science and Analytics for Smart Sustainable Urbanism. Advances in
Science, Technology &Innovation, Springer, Cham, 2019.
[41] A. Singh, A.R. Singla, Constructing definition of smart cities from systems
thinking view, Kybernetes 50 (6) (2021) 1919e1950, https://doi.org/
10.1108/K-05-2020-0276.
[42] A.M. Toli, N. Murtagh, The concept of sustainability in smart city definitions, .
Frontiers in built environment, 6(77), https://doi.org/10.3389/fbuil.2020.
00077, 2020.
[43] M. Dornh€
ofer, C. Weber, J. Zenkert, M. Fathi, A data-driven smart city
transformation model utilizing the green knowledge management cube, in:
2019 IEEE International Smart Cities Conference (ISC2), 2019.
[44] Smart cities for sustainable development, in: R.K. Mishra, C.L. Kumari,
S. Chachra, P.S.J. Krishna, A. Dubey, R.B. Singh (Eds.), Advances in
Geographical and Environmental Sciences, Springer, Singapore, 2022,
https://doi.org/10.1007/978-981-16-7410-5_1.
[45] M.K. Sutherland, M.E. Cook, Data-driven smart cities: a closer look at orga-
nizational, technical and data complexities, in: Proceedings of the 18th
Annual International Conference on Digital Government Research, 2017,
https://doi.org/10.1145/3085228.3085239.
[46] Anna Visvizi, Raquel Perez del Hoyo, Smart Cities and the UN SDGs, Elsevier,
2021, https://doi.org/10.1016/C2020-0-01556-2.
[47] A.B. Haque, B. Bhushan, G. Dhiman, Conceptualizing smart city applications:
requirements, architecture, security issues, and emerging trends, Expet Syst.
39 (5) (2022) e12753.
[48] C. Manchanda, N. Sharma, R. Rathi, B. Bhushan, M. Grover, Neoteric security
and privacy sanctuary technologies in smart cities, in: 2020 IEEE 9th Inter-
national Conference on Communication Systems and Network Technologies,
CSNT), Gwalior, India, 2020, pp. 236e241, https://doi.org/10.1109/
CSNT48778.2020.9115780.
[49] R. Register, Ecocity Berkeley: Building Cities for a Healthy Future, North
Atlantic Books, 1987.
[50] S.E. Bibri, Data-driven smart sustainable cities of the future: new conceptions
of and approaches to the spatial scaling of urban form, Future Cities Environ.
7 (1) (2021) 4, https://doi.org/10.5334/fce.120.
[51] M. De Jong, S. Joss, D. Schraven, C. Zhan, M. Weijnen, Sustainable-smart-
resilient low carbon-eco-knowledge cities; making sense of a multitude of
concepts promoting sustainable urbanization, J. Clean. Prod. 109 (2015)
25e38.
[52] E. Rapoport, A.-L. Vernay, Defining the eco-city: a discursive approach, in:
Management and Innovation for a Sustainable Built Environment Confer-
ence, International Eco-Cities Initiative, Amsterdam, The Netherlands, 2011.
[53] S.E. Bibri, J. Krogstie, Environmentally data-driven smart sustainable cities:
applied innovative solutions for energy efficiency, pollution reduction, and
urban metabolism, Energy Informat. 3 (1) (2020) 29, https://doi.org/10.1186/
s42162-020-00130-8.
[54] F. Caprotti, Smart to green: smart eco-cities in the green economy, in: The
Routledge Companion to Smart Cities, Routledge, 2020, pp. 200e209.
[55] M. Tan Mullins, A. Cheshmehzangi, S. Chien, L. Xie, Smart-eco Cities in China:
Trends and City Profiles 2016, University of Exeter, 2017.
[56] P. Sp€
ath, T. Hawxwell, R. John, S. Li, E. L€
offler, V. Riener, S. Utkarsh, Smart-eco
Cities in Germany: Trends and City Profiles, University of Exeter Press, 2017.
[57] S.E. Bibri, J. Krogstie, Smart eco-city strategies and solutions for sustain-
ability: the cases of royal seaport, Stockholm, and Western Harbor, Malm€
o,
Sweden, Urban Sci. 4 (2020) 11, https://doi.org/10.3390/urbansci4010011,
2020.
[58] R. Ghosh, D. Sengupta, Smart urban metabolism: a big-data and machine
learning perspective, in: R. Bhadouria, S. Tripathi, P. Singh, P.K. Joshi, R. Singh
(Eds.), Urban Metabolism and Climate Change, Springer, Cham, 2023, https://
doi.org/10.1007/978-3-031-29422-8_16.
[59] S.E. Bibri, The IoT and big data analytics for smart sustainable cities: enabling
technologies and practical applications, in: Advances in the Leading Para-
digms of Urbanism and Their Amalgamation: Compact Cities, EcoeCities, and
DataeDriven Smart Cities, 2020, pp. 191e226.
[60] D. Poole, A. Macworth, R. Goebel, Computational Intelligence: A Logical
Approach, Oxford University Press, New York, 1998.
[61] S.J. Russell, P. Norvig, Artificial Intelligence: A Modern Approach, Pearson
Education Limited, 2016.
[62] A. Brabazon, M. O'Neill, S. McGarraghy, Natural Computing Algorithms, vol.
554, Springer, Berlin, 2015.
[63] A. Kaplan, M. Haenlein, Siri, Siri, in my hand: who's the fairest in the land?
On the interpretations, illustrations, and implications of artificial intelli-
gence, Bus. Horiz. 62 (1) (2019) 15e25, https://doi.org/10.1016/
j.bushor.2018.08.004.
[64] A. Martınez-Miranda, Aldea Emotions in human and artificial intelligence
Computers in Human Behavior 21 (2) (2005) 323e341S.
[65] M. Ghallab, D. Nau, P. Traverso, Automated Planning and Acting, Cambridge
University Press, 2016.
[66] S.M. Omohundro, The Nature of Self-Improving Artificial Intelligence, Sin-
gularity Summit, 2007, 2008.
[67] Al-Sahaf, Y. Bi, Q. Chen, A. Lensen, Y. Mei, Y. Sun, et al., A survey on evolu-
tionary machine learning, J. Royal Soc. New Zealand 49 (2) (2019) 205e228.
[68] S.C. Pandey, G.C. Nandi, Convergence of knowledge, nature and computa-
tions: A review, Soft Comput. 20 (1) (2016) 319e342.
[69] N. Tyagi, B. Bhushan, Demystifying the role of Natural Language Processing
(NLP) in smart city applications: background, motivation, recent advances,
and future research directions, Wireless Pers. Commun. 130 (2023)
857e908, https://doi.org/10.1007/s11277-023-10312-8.
[70] S. Baker, W. Xiang, Artificial intelligence of things for smarter healthcare: a
survey of advancements, challenges, and opportunities, IEEE Commun. Surv.
Tutorials (2023), https://doi.org/10.1109/COMST.2023.3256323.
[71] S.E. Bibri, The metaverse as a virtual model of platform urbanism: its
converging AIoT, XReality, neurotech, and nanobiotech and their applica-
tions, challenges, and risks, Smart Cities 6 (3) (2023) 1345e1384, https://
doi.org/10.3390/smartcities6030065.
[72] Z. Ullah, F. Al-Turjman, L. Mostarda, R. Gagliardi, Applications of artificial
intelligence and machine learning in smart cities, Comput. Commun. 154
(2020) 313e323.
[73] C.G. García, E.R. Nú~
nez Vald
ez, V. García Díaz, B.C. Pelayo García-Bustelo,
J.M. Cueva Lovelle, A review of artificial intelligence in the internet of things,
Int. J. Interact. Multimed. Artif. Intell. 5 (2019).
[74] T.S. Brisimi, T. Mela Chen, A. Olshevsky, I.C. Paschalidis, W. Shi, Federated
learning of predictive models from federated electronic health records, Int. J.
Med. Informat. 112 (Apr. 2018) 59e67.
[75] F. Zhuang, et al., A comprehensive survey on transfer learning, Proc. IEEE,
early access, Jul. 7 (2020) 2020, https://doi.org/10.1109/
JPROC.2020.3004555.
[76] Y Q. Yao Wang, J.T. Kwok, L.M. Ni, Generalizing from a few examples: a
survey on few-shot learning, ACM Comput. Surv. 53 (3) (2020) 1e34.
[77] I.U. Din, M. Guizani, J.J.P.C. Rodrigues, S. Hassan, V.V. Korotaev, Machine
learning in the internet of things: designed techniques for smart cities,
Future Generat. Comput. Syst. 100 (2019) 826e843.
[78] A. Mosavi, M. Salimi, S. Faizollahzadeh Ardabili, T. Rabczuk, S. Shamshirband,
A.R. Varkonyi-Koczy, State of the art of machine learning models in energy
systems, Syst. Rev. 12 (7) (2019) 1301. https://www.mdpi.com/1996-1073/
12/7/1301.
[79] T.M. Mitchell, Machine Learning, McGraw-Hill Science/Engineering/Math,
1997.
[80] U. Khalil, O.A. Malik, M. Uddin, C.-L. Chen, A comparative analysis on
blockchain versus centralized authentication architectures for IoT-enabled
smart devices in smart cities: a comprehensive review, recent advances,
and future research directions, Sensors 22 (2022) 5168, https://doi.org/
10.3390/s22145168.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
26
[81] Sita Rani, Raman Kumar, Bibliometric review of actuators: key automation
technology in a smart city framework, Mater. Today: Proc. 60 (3) (2022),
https://doi.org/10.1016/j.matpr.2021.12.469.
[82] M.N. Akhter, S. Mekhilef, H. Mokhlis, N. Mohamed Shah, Review on fore-
casting of photovoltaic power generation based on machine learning and
metaheuristic techniques, 13(7), 1009-1023, https://doi.org/10.1049/iet-rpg.
2018.5649, 2019.
[83] U.K. Das, K.S. Tey, M. Seyedmahmoudian, S. Mekhilef, M.Y.I. Idris, W. Van
Deventer, B. Horan, A. Stojcevski, Forecasting of photovoltaic power gener-
ation and model optimization: a review, Renew. Sustain. Energy Rev. 81
(2018) 912e928, https://doi.org/10.1016/j.rser.2017.08.017.
[84] S. Khan, D. Paul, P. Momtahan, M. Aloqaily, Artificial intelligence framework
for smart city microgrids: state of the art, challenges, and opportunities, in:
2018 Third International Conference on Fog and Mobile Edge Computing
(FMEC), 2018.
[85] S.K. Jha, J. Bilalovic, A. Jha, N. Patel, H. Zhang, Renewable energy: present
research and future scope of Artificial Intelligence, Renew. Sustain. Energy
Rev. 77 (2017) 297e317, https://doi.org/10.1016/j.rser.2017.04.018.
[86] D. Mhlanga, Artificial intelligence and machine learning for energy con-
sumption and production in emerging markets: a review, Energies 16 (2023)
745, https://doi.org/10.3390/en16020745.
[87] S. Su, X. Yan, K. Agbossou, R. Chahine, Y. Zong, Artificial intelligence for
hydrogen-based hybrid renewable energy systems: a review with case
study, J. Phys. Conf. 2208 (2022) 012013.
[88] Alsadi, T. Khatib Photovoltaic power systems optimization research status: A
review of criteria, constrains, models, techniques, and software tools. Appl.
Sci., 8 (10) (2018), p. 1761.
[89] E. Bertone, M.A. Burford, D.P. Hamilton, Fluorescence probes for real-time
remote cyanobacteria monitoring: a review of challenges and opportu-
nities, Water Res. 141 (2018) 152e162, https://doi.org/10.1016/
j.watres.2018.05.001.
[90] O. Oyebode, D. Stretch, Neural network modeling of hydrological systems: A
review of implementation techniques, Nat. Res. Model. 32 (1) (2019) e12189.
[91] C. Shen, A transdisciplinary review of deep learning research and its rele-
vance for water resources scientists, Water Resour. Res. 54 (11) (2018)
8558e8593, https://doi.org/10.1029/2018WR022643.
[92] H. Tyralis, G. Papacharalampous, A. Langousis, A brief review of random
forests for water scientists and practitioners and their recent history in water
resources, 11(5), 910, https://www.mdpi.com/2073-4441/11/5/910, 2019.
[93] Sakti Prajna Mahardhika, Okkie Putriani, Deployment and use of Artificial
Intelligence (AI) in water resources and water management, IOP Conf. Ser.
Earth Environ. Sci. 1195 (2023) 012056.
[94] B. Fang, J. Yu, Z. Chen, A.I. Osman, M. Farghali, I. Ihara, et al., Artificial in-
telligence for waste management in smart cities: a review, Environ. Chem.
Lett. (2023), https://doi.org/10.1007/s10311-023-01604-3.
[95] I. Nasir, G.A. Aziz Al-Talib, Waste classification using artificial intelligence
techniques: literature review, Technium: Roman. J. Appl. Sci. Technol. 5
(2023) 49e59, https://doi.org/10.47577/technium.v5i.8345.
[96] Adnane Mounadel, Hamid Ech-Cheikh, Sa^
ad Lissane Elhaq, Rachid, Ahmed,
Sadik Mohamed, Abdellaoui Bilal, Application of artificial intelligence tech-
niques in municipal solid waste management: a systematic literature review,
Environ. Technol. Rev. 12 (1) (2023) 316e336, https://doi.org/10.1080/
21622515.2023.2205027.
[97] S. Salcedo-Sanz, L. Cuadra, M.J.A. Vermeij, A review of Computational Intel-
ligence techniques in coral reef-related applications, Ecol. Inf. 32 (2016)
107e123, https://doi.org/10.1016/j.ecoinf.2016.01.008.
[98] T. Toivonen, V. Heikinheimo, C. Fink, A. Hausmann, T. Hiippala, O. J€
arv, et al.,
Social media data for conservation science: a methodological overview, Biol.
Conserv. 233 (2019) 298e315.
[99] E.K. Nti, S.J. Cobbina, E.E. Attafuah, E. Opoku, M.A. Gyan, Environmental
sustainability technologies in biodiversity, energy, transportation and water
management using artificial intelligence: a systematic review, Sust. Fut. 4
(2022) 100068.
[100] R.L. Abduljabbar, H. Dia, S. Liyanage, S.A. Bagloee, Applications of artificial
intelligence in transport: an overview, Sustainability 11 (1) (2019) 189,
https://doi.org/10.3390/su11010189.
[101] S. Liyanage, H. Dia, R. Abduljabbar, S.A. Bagloee, Flexible mobility on-
demand: an environmental scan, Sustainability 11 (5) (2019) 1262.
[102] W.J. Raseman, J.R. Kasprzyk, F.L. Rosario-Ortiz, J.R. Stewart, B. Livneh,
Emerging investigators series: a critical review of decision support systems
for water treatment: making the case for incorporating climate change and
climate extremes, Environ. Sci.: Water Res. Technol. 3 (1) (2017) 18e36.
[103] T.-H. Yang, W.-C. Liu, A general overview of the risk-reduction strategies for
floods and droughts, Sustainability 12 (7) (2020) 2687.
[104] L. Chen, Z. Chen, Y. Zhang, et al., Artificial intelligence-based solutions for
climate change: a review, Environ. Chem. Lett. (2023), https://doi.org/
10.1007/s10311-023-01617-y.
[105] L. Chen, Z. Chen, Y. Zhang, et al., Artificial intelligence-based solutions for
climate change: a review, Environ. Chem. Lett. (2023), https://doi.org/
10.1007/s10311-023-01617-y.
[106] Z. Allam, A.Z. Dhunny, On big data, artificial intelligence and smart cities,
Cities 89 (2019) 80e91.
[107] P.J. Navarathna, V.P. Malagi, Artificial Intelligence in Smart City Analysis
International Conference on Smart Systems Inventive Technology, ICSSIT,
2018, pp. 44e47, 2018, https://doi:10.1109/ICSSIT.2018.8748476.
[108] S.C. Mukhopadhyay, S.K.S. Tyagi, N.K. Suryadevara, et al., Artificial
intelligence-based sensors for next generation IoT applications: a review,
IEEE Sensor. J. 21 (2021) 24920e24932.
[109] A. Dresch, D.P. Lacerda, J.A.V. Antunes Jr., Design Science Research: A Method
for Science and Technology Advancement, Springer International Publishing,
Heidelberg, 2015.
[110] S. Seuring, S. Gold, Conducting content-analysis based literature reviews in
supply chain management, Supply Chain Manag. 17 (5) (2012) 544e555,
https://doi.org/10.1108/13598541211258609.
[111] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P.G. The, Preferred reporting
items for systematic reviews and meta-analyses: the PRISMA Statement,
PLoS Med. 6 (7) (2009) e1000097, https://doi.org/10.1371/
journal.pmed.1000097.
[112] Page M.J., McKenzie J.E., Bossuyt P.M., Boutron I., Hoffmann T.C., Mulrow C.D.,
Shamseer L., Tetzlaff J.M., Akl E.A., Brennan S.E., Chou .R, Glanville J., Grim-
shaw J.M., Hr
objartsson A., Lalu M.M., Li T., Loder E.W., Mayo-Wilson E.,
McDonald S. et al (2021) The PRISMA 2020 statement: an updated guideline
for reporting systematic reviews. BMJ 372:n71. https://doi.org/10.1136/bmj.
n71.
[113] G. Talan, G.D. Sharma, Doing well by doing good: a systematic review and
research agenda for sustainable investment, Sustainability 11 (2) (2019).
[114] D. Gough, S. Oliver, J. Thomas, An Introduction to Systematic Reviews, SAGE
Publications, 2017.
[115] H.M. Cooper, Research Synthesis and Meta-Analysis: A Step-by-step
Approach, Sage Publications, 2017.
[116] M. Dixon-Woods, S. Agarwal, D. Jones, B. Young, A. Sutton, Synthesizing
qualitative and quantitative evidence: A review of possible methods,
J. Health Serv. Res. Policy 10 (1) (2005) 45e53.
[117] J. Popay, H. Roberts, A. Sowden, M. Petticrew, L. Arai, M. Rodgers, et al.,
Guidance on the Conduct of Narrative Synthesis in Systematic Reviews. A
Product from the ESRC Methods Programme, Lancaster University, 2006.
[118] Karolina Lisy, Kylie Porritt, Narrative synthesis: considerations and chal-
lenges, Int. J. Evid. Base. Healthc. 14 (4) (December 2016) 201, https://
doi.org/10.1097/01.XEB.0000511348.97198.8c.
[119] H. Chourabi, T. Nam, S. Walker, J.R. Gil-Garcia, S. Mellouli, K. Nahon,
T.A. Pardo, H.J. Scholl, Understanding smart cities: an integrative framework,
in: 45th Hawaii International Conference on System Sciences, 2012, https://
doi.org/10.1109/hicss.2012.615.
[120] L.M. Correia, K. Wuenstel, Smart Cities Applications and Requirements
[White Paper](White Paper, Issue. Net!Works, 2011.
[121] L. Atzori, A. Iera, G. Morabito, The internet of things: a survey, Comput.
Network. 54 (15) (2010) 2787e2805, https://doi.org/10.1016/
j.comnet.2010.05.010.
[122] R. Tiwari, R. Cervero, L. Schipper, Driving CO2 reduction by integrating
transport and urban design strategies, Cities 28 (5) (2011) 394e405, https://
doi.org/10.1016/j.cities.2011.05.005.
[123] A. Caragliu, Smart Cities in Europe 3 Rd Central European Conference in
Regional Saence-CERS, 7-9 October, 2009, Ko
sice, Slovak Republic, 2009.
[124] R. Giffinger, C. Fertner, H. Kramar, R. Kalasek, N. Pichler-Milanovic, E. Meijers,
Smart Cities: Ranking of European Medium-Sized Cities, Centre of Regional
Science, Vienna University of Technology, Vienna, Austria, 2007. Available
online: http://www.smart-cities.eu/download/smart_cities_final_report.pdf.
[125] D. Toppeta, The Smart City Vision: How Innovation and ICT Can Build
Smart,“livable”, Sustainable Cities, vol. 5, The innovation knowledge foun-
dation, 2010, pp. 1e9.
[126] M. Fujii, T. Fujita, X. Chen, et al., Smart recycling of organic solid wastes in an
environmentally sustainable society, Resour. Conserv. Recycl. 63 (2012) 1e8.
[127] V. Sempere-Pay
a, S. Santonja-Climent, Integrated sensor and management
system for urban waste water networks and prevention of critical situations,
Comput. Environ. Urban Syst. 36 (2012) 65e80.
[128] M. Batty, K.W. Axhausen, F. Giannotti, A. Pozdnoukhov, A. Bazzani,
M. Wachowicz, G. Ouzounis, Y. Portugali, Smart cities of the future, Eur. Phys.
J. Spec. Top. 214 (1) (2012) 481e518, https://doi.org/10.1140/epjst/e2012-
01703-3.
[129] K. Su, J. Li, H. Fu, Smart city and the applications, in: International Conference
on Electronics, Communications and Control (ICECC), IEEE(2011), 2011,
pp. 1028e1031.
[130] E. Al Nuaimi, H. Al Neyadi, N. Mohamed, J. Al-Jaroodi, Applications of big data
to smart cities, J. Internet Services Appl. 6 (1) (2015) 25, https://doi.org/
10.1186/s13174-015-0041-5.
[131] L.M.A. Bettencourt, The Uses of Big Data in Cities, Santa Fe Institute, 2014.
[132] L. Deren, C. Jianjun, Y. Yao, Big data in smart cities, Sci. China Inf. Sci. 58
(2015) 1e12.
[133] H.M. Kim, S.S. Han, City profile: Seoul, Cities 29 (2) (2012) 142e154.
[134] Z. Ji, I. Ganchev, M. O'Droma, L. Zhao, X. Zhang, A cloud-based car parking
middleware for IoT-based smart cities: design and implementation, Sensors
14 (12) (2014) 22372e22393.
[135] D. Kyriazis, T. Varvarigou, D. White, A. Rossi, J. Cooper, Sustainable smart city
IoT applications: heat and electricity management &Eco-conscious cruise
control for public transportation, in: 2013 IEEE 14th International Sympo-
sium on" A World of Wireless, Mobile and Multimedia Networks"(WoW-
MoM), IEEE, 2013, pp. 1e5.
[136] A. Zanella, N. Bui, A. Castellani, L. Vangelista, M. Zorzi, Internet of things for
smart cities, IEEE Internet Things J. 1 (1) (2014) 22e32.
[137] I.A.T. Hashem, V. Chang, N.B. Anuar, K. Adewole, I. Yaqoob, A. Gani, E. Ahmed,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
27
H. Chiroma, The role of big data in smart city, Int. J. Inf. Manag. 36 (5) (2016)
748e758.
[138] K. Nikitin, N. Lantsev, A. Nugaev, A. Yakavleva, Data-driven cities: from
concept to applied solutions, PricewaterhouseCoopers (PWC) (2016).
[139] M.M. Rathore, A. Ahmad, A. Paul, S. Rho, Urban planning and building smart
cities based on the Internet of Things using Big Data analytics, Comput.
Network. 101 (2016) 63e80.
[140] A.J. Sinaeepourfard, X.M.-B. Garcia, E. Marín-Tordera, J. Cirera, G. Grau,
F. Casaus, Estimating smart city sensors data generation current and future
data in the city of Barcelona, in: Proceedings of Conference: the 15th IFIP
Annual Mediterranean Ad Hoc Networking Workshop, 2016.
[141] S. Ameer, M.A. Shah, Exploiting big data analytics for smart urban planning,
in: 2018 IEEE 88th Vehicular Technology Conference (VTC-Fall), IEEE, 2018,
August, pp. 1e5.
[142] B. Cheng, G. Solmaz, F. Cirillo, E. Kovacs, K. Terasawa, A. Kitazawa, FogFlow:
easy programming of IoT services over cloud and edges for smart cities, IEEE
Internet Things J. 5 (2) (2018) 696e707.
[143] C. Perera, Y. Qin, J.C. Estrella, S. Reiff-Marganiec, A.V. Vasilakos, Fog
computing for sustainable smart cities: a survey, ACM Comput. Surv. 50 (3)
(2017) 1e43.
[144] M.M. Rathore, A. Paul, W.H. Hong, H. Seo, I. Awan, S. Saeed, Exploiting IoT
and big data analytics: defining smart digital city using real-time urban data,
Sustain. Cities Soc. 40 (2018) 600e610.
[145] B.N. Silva, M. Khan, K. Han, Toward sustainable smart cities: a review of
trends, architectures, components, and open challenges in smart cities,
Sustain. Cities Soc. 38 (2018) 697e713, https://doi.org/10.1016/
j.scs.2018.01.053.
[146] R. Kitchin, The ethics of smart cities and urban science, Phil. Trans. Math.
Phys. Eng. Sci. 374 (2083) (2016) 1e15, https://doi.org/10.1098/
rsta.2016.0115.
[147] M. H€
ojer, J. Wangel, Smart sustainable cities: definition and challenges. ICT
innovations for sustainability, Adv. Intell. Syst. Comput. (2015) 333e349.
[148] A. Kramers, M. H€
ojer, N. L€
ovehagen, J. Wangel, Smart sustainable cities e
exploring ICT solutions for reduced energy use in cities, Environ. Model.
Software 56 (2014) 52e62, https://doi.org/10.1016/j.envsoft.2013.12.019.
[149] H. Shahrokni, F. Levihn, N. Brandt, Big meter data analysis of the energy
efficiency potential in Stockholm's building stock, Energy Build. 78 (2014)
153e164, https://doi.org/10.1016/j.enbuild.2014.04.017.
[150] S. Shahrokni, H. Bram Van der, L. David, B. Nils, Big data GIS analytics toward
efficient waste management in Stockholm, in: Proceedings of the 2014
Conference ICT for Sustainability, 2014, https://doi.org/10.1016/
j.ijinfomgt.2016.05.002.
[151] H. Shahrokni, D. lazarevic, N. brandt, Smart urban metabolism: toward a
real-time understanding of the energy and material flows of a city and its
citizens, J. Urban Technol. 22 (1) (2015) 65e86.
[152] F. Caprotti, R. Cowley, A. Flynn, S. Joss, L. Yu, Smart-Eco Cities in the UK:
Trends and CityProfiles 2016, Exeter: University of Exeter (SMART-ECO
Project), 2016.
[153] O. Pasichnyi, F. Levihn, H. Shahrokni, J. Wallin, O. Kordas, Data-driven stra-
tegic planning of building energy retrofitting: the case of Stockholm, J. Clean.
Prod. 233 (2019) 546e560.
[154] N. Noori, T. Hoppe, M. de Jong, Classifying pathways for smart city devel-
opment: comparing design, governance and implementation in Amsterdam,
Barcelona, Dubai, and Abu Dhabi, Sustainability 12 (10) (2020) 4030.
[155] A. Kramers, J. Wangel, M. H€
ojer, Governing the smart sustainable city: the
case of the Stockholm Royal Seaport, ICT Sustain. 46 (2016) 99e108, 2016.
[156] O. Pasichnyi, J. Wallin, F. Levihn, H. Shahrokni, O. Kordas, Energy perfor-
mance certificatesdnew opportunities for data-enabled urban energy policy
instruments? Energy Pol. 127 (2019) 486e499.
[157] S.E. Bibri, Unprecedented innovations in sustainable urban planning: novel
analytical solutions and data-driven decision-making processes, in: Smart
Sustainable Cities of the Future. The Urban Book Series, Springer, Cham,
2018, https://doi.org/10.1007/978-3-319-73981-6_5.
[158] A.L. Balogun, D. Marks, R. Sharma, H. Shekhar, C. Balmes, D. Maheng,
A. Arshad, P. Salehi, Assessing the potentials of digitalisation as a tool for
climate change adaptation and sustainable development in urban centres,
Sustain. Cities Soc. 53 (2020) 101888.
[159] Y.K. Dwivedi, L. Hughes, A.K. Kar, A.M. Baabdullah, P. Grover, R. Abbas, et al.,
Climate change and COP26: are digital technologies and information man-
agement part of the problem or the solution? An editorial reflection and call
to action, Int. J. Inf. Manag. 63 (2022) 102456.
[160] J. Falk, O. Gaffney, A.K. Bhowmik, P. Bergmark, V. Galaz, N. Gaskell, et al.,
Exponential Roadmap V 1.5. 1, Future Earth, Stockholm, Sweden, 2020.
[161] M. Demirci, F. Ünes¸ , S. K €
orü, Modeling of groundwater level using artificial
intelligence techniques: a case study of Reyhanli region in Turkey, Appl. Ecol.
Environ. Res. 17 (2019) 2651e2663, https://doi.org/10.15666/aeer/1702_
26512663.
[162] A.K. Poul, M. Shourian, H. Ebrahimi, A Comparative Study of MLR, KNN, ANN,
and ANFIS Models with Wavelet Transform in Monthly Stream Flow Pre-
diction, Water Resources Management, 2019, pp. 1e17.
[163] M. Al-Mukhtar, F. Al-Yaseen, Modeling water quality parameters using data-
driven models, a case study Abu-Ziriq Marsh in South of Iraq, Hydrology 6
(1) (2019) 24, https://doi.org/10.3390/hydrology6010024.
[164] M. Najafzadeh, A. Ghaemi, Prediction of the five-day biochemical oxygen
demand and chemical oxygen demand in natural streams using machine
learning methods, Environ. Monit. Assess. 191 (6) (2019) 380, https://
doi.org/10.1007/s10661-019-7446-8.
[165] M.H. Calp, A hybrid anfis-ga approach for estimation of regional rainfall
amount, J. Sci. 32 (2019) 145e162.
[166] L. Ferreira, A. Duarte, F. Cunha, E. Fernandes-Filho, Multivariate adaptive
regression splines (MARS) applied to daily reference evapotranspiration
modeling with limited weather data, Acta Sci. Agron. 41 (2019), https://
doi.org/10.4025/actasciagron.v41i1.39880.
[167] I. Rojek, J. Studzinski, Detection and localization of water leaks in water nets
supported by an ICT system with artificial intelligence methods as a way
forward for smart cities, Sustainablity vol. 11 (2019) 518, 2.
[168] J.
A. Gonz
alez Ordiano, S. Waczowicz, V. Hagenmeyer, R. Mikut, Energy
forecasting tools and services, WIREs Data Min. Knowledge Discovery 8 (2)
(2018) e1235, https://doi.org/10.1002/widm.1235.
[169] Y.L. Merizalde, O. Hern
andez-Callejo, V. Duque-Perez, Alonso-G
omez main-
tenance models applied to wind turbines A comprehensive overview, En-
ergies 12 (2) (2019) 225.
[170] H. Sellak, B. Ouhbi, B. Frikh, I. Palomares, Toward next-generation energy
planning decision-making: an expert-based framework for intelligent deci-
sion support, Renew. Sustain. Energy Rev. 80 (2017) 1544e1577, https://
doi.org/10.1016/j.rser.2017.07.013.
[171] F. Alrukaibi, R. Alsaleh, T. Sayed, Applying machine learning and statistical
approaches for travel time estimation in partial network coverage, 11(14),
3822, https://doi.org/10.3390/su11143822, 2019.
[172] T. Zhou, G. Han, X. Xu, C. Han, Y. Huang, J. Qin, A learning-based multimodel
integrated framework for dynamic trafficflow forecasting, Neural Process.
Lett. 49 (1) (2019) 407e430, https://doi.org/10.1007/s11063-018-9804-x.
[173] N. Abdelkafi,K.T
€
auscher, Business models for sustainability from a system
dynamics perspective, Organ. Environ. 29 (1) (2016) 74e96.
[174] Flah Aymen, Mahmoudi Chokri, Sbita Lassaad, A novel energy optimization
approach for electrical vehicles in a smart city, Energies 12 (5) (2019) 929,
https://doi.org/10.20944/preprints201901.0214.v1.
[175] K.J. Bagstad, J.M. Reed, D.J. Semmens, B.C. Sherrouse, A. Troy, Linking bio-
physical models and public preferences for ecosystem service assessments: a
case study for the Southern Rocky Mountains, Reg. Environ. Change 16 (7)
(2016) 2005e2018, https://doi.org/10.1007/s10113-015-0756-7.
[176] C. Rodríguez-Soto, A. Velazquez, O. Monroy-Vilchis, P. Lemes, R. Loyola, Joint
ecological, geographical and cultural approach to identify territories of op-
portunity for large vertebrates conservation in Mexico, Biodivers. Conserv.
26 (8) (2017) 1899e1918.
[177] K. Sharps, D. Masante, A. Thomas, B. Jackson, J. Redhead, L. May, H. Prosser,
B. Cosby, B. Emmett, L. Jones, Comparing strengths and weaknesses of three
ecosystem services modelling tools in a diverse UK river catchment, Sci.
Total Environ. 584e585 (2017) 118e130, https://doi.org/10.1016/
j.scitotenv.2016.12.160.
[178] S. Willcock, J. Martínez-L
opez, D.A.P. Hooftman, K.J. Bagstad, S. Balbi,
A. Marzo, et al., Machine learning for ecosystem services, Ecosyst. Serv. 33
(2018) 165e174, https://doi.org/10.1016/j.ecoser.2018.04.004.
[179] A.-b. Zhang, M.-d. Hao, C.-q. Yang, Z.-y. Shi, BarcodingR: an integrated r
package for species identification using DNA barcodes, 8(5), 627-634,
https://doi.org/10.1111/2041-210X.12682, 2017.
[180] A.D. Mehr, V. Nourani, E. Kahya, B. Hrnjica, A.M. Sattar, Z.M. Yaseen, Genetic
programming in water resources engineering: a state-of-the-art review,
J. Hydrol. 566 (2018) 643e667.
[181] B.B. Sahoo, R. Jha, A. Singh, D. Kumar, Application of support vector regres-
sion for modeling low flow time series, KSCE J. Civ. Eng. 23 (2) (2019)
923e934.
[182] N. Valizadeh, M. Mirzaei, M.F. Allawi, H.A. Afan, N.S. Mohd, A. Hussain, A. El-
Shafie, Artificial intelligence and geo-statistical models for stream-flow
forecasting in ungauged stations: state of the art, Nat. Hazards 86 (3)
(2017) 1377e1392, https://doi.org/10.1007/s11069-017-2740-7.
[183] S. Alsadi, T. Khatib, Photovoltaic power systems optimization research sta-
tus: a review of criteria, Constrain. Models Techn. Software Tools 8 (10)
(2018) 1761, https://doi.org/10.3390/app8101761.
[184] S.M. Dawoud, X. Lin, M.I. Okba, Hybrid renewable microgrid optimization
techniques: a review, Renew. Sustain. Energy Rev. 82 (2018) 2039e2052,
https://doi.org/10.1016/j.rser.2017.08.007.
[185] A. Youssef, M. El-Telbany, A. Zekry, The role of artificial intelligence in photo-
voltaic systems design and control: a review, Renew. Sustain. Energy Rev. 78
(2017) 72e79, https://doi.org/10.1016/j.rser.2017.04.046.
[186] Z. Wang, R.S. Srinivasan, A review of artificial intelligence based building
energy use prediction: contrasting the capabilities of single and ensemble
prediction models, Renew. Sustain. Energy Rev. 75 (2017) 796e808, https://
doi.org/10.1016/j.rser.2016.10.079.
[187] W. Jiang, L. Zhang, Geospatial data to images: a deep-learning framework for
traffic forecasting, Tsinghua Sci. Technol. 24 (1) (2019) 52e64, https://
doi.org/10.26599/TST.2018.9010033.
[188] L.V. Kalmykov, V.L. Kalmykov, A solution to the biodiversity paradox by
logical deterministic cellular automata, Acta Biotheor. 63 (2) (2015)
203e221, https://doi.org/10.1007/s10441-015-9257-9.
[189] V. Ochoa, N. Urbina-Cardona, Tools for spatially modeling ecosystem ser-
vices: publication trends, conceptual reflections and future challenges,
Ecosyst. Serv. 26 (2017) 155e169.
[190] B.D. Bowes, J.M. Sadler, M.M. Morsy, M. Behl, J.L. Goodall, Forecasting
groundwater table in a flood prone coastal city with long short-term
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
28
memory and recurrent neural networks, 11(5), 1098, https://doi.org/10.
3390/w11051098, 2019.
[191] D.J. Jeon, S.J. Ki, Y. Cha, Y. Park, J.H. Kim, New methodology of evaluation of
best management practices performances for an agricultural watershed ac-
cording to the climate change scenarios: a hybrid use of deterministic and
decision support models, Ecol. Eng. 119 (2018) 73e83, https://doi.org/
10.1016/j.ecoleng.2018.05.006.
[192] A. Ullah, N. Salehnia, S. Kolsoumi, A. Ahmad, T. Khaliq, Prediction of effective
climate change indicators using statistical downscaling approach and impact
assessment on pearl millet (Pennisetum glaucum L.) yield through Genetic
Algorithm in Punjab, Pakistan, Ecol. Indicat. 90 (2018) 569e576, https://
doi.org/10.1016/j.ecolind.2018.03.053.
[193] M. Skiba, M. Mr
owczy
nska, A. Bazan-Krzywosza
nska, Modeling the eco-
nomic dependence between town development policy and increasing en-
ergy effectiveness with neural networks. Case study: the town of Zielona
G
ora, Appl. Energy 188 (2017) 356e366, https://doi.org/10.1016/
j.apenergy.2016.12.006.
[194] G.S. Atsalakis, Using computational intelligence to forecast carbon prices,
Appl. Soft Comput. 43 (2016) 107e116.
[195] M. Raji, A. Dashti, P. Amani, A.H. Mohammadi, Efficient estimation of CO2
solubility in aqueous salt solutions, J. Mol. Liq. 283 (2019) 804e815.
[196] T.H. Woo, Global warming analysis for greenhouse gases impacts compara-
ble to carbon-free nuclear energy using neuro-fuzzy algorithm, 17(2), 219-
233, https://doi.org/10.1504/ijgw.2019.097862, 2019.
[197] M. Mokhtarzad, F. Eskandari, N. Jamshidi Vanjani, A. Arabasadi, Drought
forecasting by ANN, ANFIS, and SVM and comparison of the models, Environ.
Earth Sci. 76 (21) (2017) 729, https://doi.org/10.1007/s12665-017-7064-0.
[198] R. Shen, A. Huang, B. Li, J. Guo, Construction of a drought monitoring model
using deep learning based on multi-source remote sensing data, Int. J. Appl.
Earth Obs. Geoinf. 79 (2019) 48e57, https://doi.org/10.1016/
j.jag.2019.03.006.
[199] A. Piasecki, J. Jurasz, J.F. Adamowski, Forecasting surface water-level fluc-
tuations of a small glacial lake in Poland using a wavelet-based artificial
intelligence method, Acta Geophys. 66 (5) (2018) 1093e1107.
[200] T. Ise, Y. Oba, Forecasting climatic trends using neural networks: an exper-
imental study using global historical data [original research], 6, 32, https://
doi.org/10.3389/frobt.2019.00032, 2019.
[201] Z. Yin, Q. Feng, L. Yang, R.C. Deo, X. Wen, J. Si, S. Xiao, Future projection with
an extreme-learning machine and support vector regression of reference
evapotranspiration in a mountainous Inland watershed in North-West
China, 9(11), 880, https://doi.org/10.3390/w9110880, 2017.
[202] B. Choubin, M. Borji, A. Mosavi, F. Sajedi-Hosseini, V.P. Singh,
S. Shamshirband, Snow avalanche hazard prediction using machine learning
methods, J. Hydrol. 577 (2019) 123929, https://doi.org/10.1016/
j.jhydrol.2019.123929.
[203] L. Ji, Z. Wang, M. Chen, S. Fan, Y. Wang, Z. Shen, How much can AI techniques
improve surface air temperature forecast? da report from AI challenger
2018 global weather forecast contest, J. Meteorol. Res. 33 (5) (2019)
989e992, https://doi.org/10.1007/s13351-019-9601-0.
[204] H. Lu, H. Li, T. Liu, Y. Fan, Y. Yuan, M. Xie, X. Qian, Simulating heavy metal
concentrations in an aquatic environment using artificial intelligence models
and physicochemical indexes, Sci. Total Environ. 694 (2019) 133591, https://
doi.org/10.1016/j.scitotenv.2019.133591.
[205] P. Wang, J. Yao, G. Wang, F. Hao, S. Shrestha, B. Xue, et al., Exploring the
application of artificial intelligence technology for identification of water
pollution characteristics and tracing the source of water quality pollutants,
Sci. Total Environ. 693 (2019) 133440, https://doi.org/10.1016/
j.scitotenv.2019.07.246.
[206] Y. Ampatzidis, L. De Bellis, A. Luvisi, iPathology: robotic applications and
management of plants and plant diseases, 9(6), 1010, https://doi.org/10.
3390/su9061010, 2017.
[207] J.F. De Paz, J. Bajo, S. Rodríguez, G. Villarrubia, J.M. Corchado, Intelligent
system for lighting control in smart cities, Inf. Sci. 372 (2016) 241e255,
https://doi.org/10.1016/j.ins.2016.08.045.
[208] S. Mathur, U.S. Modani, Smart City- a gateway for artificial intelligence in
India, 5-6 March 2016, in: 2016 IEEE Students' Conference on Electrical,
Electronics and Computer Science (SCEECS), 2016, 2016, pp. 1e3, https://
doi.org/10.1109/SCEECS.2016.7509291.
[209] G. Sonetti, E. Naboni, M. Brown, Exploring the potentials of ICT tools for
human-centric regenerative design, Sustainability 10 (4) (2018) 1217,
https://doi.org/10.3390/su10041217.
[210] J. V
azquez-Canteli, S. Ulyanin, J. K€
ampf, Z. Nagy, Fusing TensorFlow with
building energy simulation for intelligent energy management in smart
cities, Sustain. Cities Soc. 45 (2018) 243e257, https://doi.org/10.1016/
j.scs.2018.11.021.
[211] E.A. Barnes, J.W. Hurrell, I. Ebert-Uphoff, C. Anderson, D. Anderson, Viewing
forced climate patterns through an AI Lens, Geophys. Res. Lett. 46 (22)
(2019) 13389e13398, https://doi.org/10.1029/2019GL084944.
[212] C. Huntingford, E.S. Jeffers, M.B. Bonsall, H.M. Christensen, T. Lees, H. Yang,
Machine learning and artificial intelligence to aid climate change research
and preparedness, Environ. Res. Lett. 14 (12) (2019) 124007.
[213] P. Chamoso, J.F. De Paz, S. Rodríguez, J. Bajo, in: H. Lindgren, J.F. De Paz,
P. Novais, A. Fern
andez-Caballero, H (Eds.), Smart Cities Simulation Envi-
ronment for Intelligent Algorithms Evaluation, 2016.
[214] K. Muhammad, J. Lloret, S.W. Baik, Intelligent and energy-efficient data
prioritization in green smart cities: current challenges and future directions,
IEEE Commun. Mag. 57 (2) (2019) 60e65, https://doi.org/10.1109/
MCOM.2018.1800371.
[215] H. Yu, Z. Yang, R. Sinnott, Decentralized big data auditing for smart city
environments leveraging blockchain technology, IEEE Access 7 (2018)
6288e6296, https://doi.org/10.1109/ACCESS.2018.2888940.
[216] K. Guo, Y. Lu, H. Gao, R. Cao, Artificial intelligence-based semantic internet of
things in a user-centric smart city, Sensors 18 (5) (2018), https://doi.org/
10.3390/s18051341.
[217] J.-h. Park, M. Salim, J. Jo, J. Sicato, S. Rathore, J. Park, CIoT-Net: a scalable
cognitive IoT based smart city network architecture, Human-centric Comput.
Info. Sci. 9 (2019) 29.
[218] S. Belkhala, S. Benhadou, K. Boukhdir, H. Medromi, Smart parking architec-
ture based on multi agent system, Int. J. Adv. Comput. Sci. Appl. 10 (2019)
378e382, https://doi.org/10.14569/IJACSA.2019.0100349.
[219] I. Antonopoulos, V. Robu, B. Couraud, et al., Artificial intelligence and ma-
chine learning approaches to energy demand-side response: a systematic
review, Renew. Sustain. Energy Rev. 130 (2020) 109899.
[220] J. Zhang, K.B. Letaief, Mobile edge intelligence and computing for the
internet of vehicles, Proc. IEEE 108 (2019) 246e261.
[221] M.K. AlOmar, M.M. Hameed, M.A. AlSaadi, Multi hours ahead prediction of
surface ozone gas concentration: robust artificial intelligence approach,
Atmos. Pollut. Res. 11 (9) (2020) 1572e1587, https://doi.org/10.1016/
j.apr.2020.06.024.
[222] M. Schultz, Artificial intelligence for air quality, Project Repository J. 12
(2022) 70e73. https://www.europeandissemination.eu/article/artificial-
intelligence-for-air-quality/18384.
[223] D. Schürholz, S. Kubler, A. Zaslavsky, Artificial intelligence-enabled context-
aware air quality prediction for smart cities, J. Clean. Prod. 271 (2020)
121941, https://doi.org/10.1016/j.jclepro.2020.121941.
[224] V. Sowmya, S. Ragiphani, Air quality monitoring system based on artificial
intelligence, in: P. Kumar Jain, Y. Nath Singh, R.P. Gollapalli, S.P. Singh (Eds.),
Advances in Signal Processing and Communication Engineering, Springer,
Singapore, 2022, p. 26, https://doi.org/10.1007/978-981-19-5550-1_26.
[225] Z. Liu, Z. Yang, M. Osmani, The relationship between sustainable built
environment, art therapy and therapeutic design in promoting health and
well-being, Int. J. Environ. Res. Publ. Health 18 (20) (2021) 10906. Retrieved
from, https://www.mdpi.com/1660-4601/18/20/10906.
[226] S.S. Amiri, S. Mottahedi, E.R. Lee, S. Hoque, Peeking inside the black-box:
explainable machine learning applied to household transportation energy
consumption, Comput. Environ. Urban Syst. 88 (2021) 101647.
[227] G. Aschwanden, J.S. Wijnands, J. Thompson, K.A. Nice, H.F. Zhao,
M. Stevenson, Learning to walk: modeling transportation mode choice dis-
tribution through neural networks, Environ. Plan. B Urban Anal. City Sci. 48
(2021) 186e199.
[228] X.M. Cheng, J.Y. Wang, H.F. Li, Y. Zhang, L. Wu, Y. Liu, A method to evaluate
task-specific importance of spatio-temporal units based on explainable
artificial intelligence, Int. J. Geogr. Inf. Sci. 35 (2020) 10.
[229] S. Fathi, R.S. Srinivasan, C.J. Kibert, R.L. Steiner, E. Demirezen, AI-based
campus energy use prediction for assessing the effects of climate change,
Sustainability 12 (2020) 3223.
[230] Z.J. Yan, B. Li, Q.Q. Li, M. Yang, An efficient multiple access control protocol
for directional dense urban traffic surveillance system, J. Intell. Transp. Syst.
24 (2020) 237e253.
[231] F.-J. Chang, L.-C. Chang, J.-F. Chen, Artificial intelligence techniques in hy-
drology and water resources management, Water 15 (10) (2023) 1846,
https://doi.org/10.3390/w15101846.
[232] P. Martínez-Santos, P. Renard, Mapping groundwater potential through an
ensemble of big data, Methods 58 (4) (2020) 583e597, https://doi.org/
10.1111/gwat.12939.
[233] S.Y. Chen, H.Y. Kuo, C. Lee, Preparing society for automated vehicles: per-
ceptions of the importance and urgency of emerging issues of governance,
regulations, and wider impacts, Sustainability 12 (2020) 7844, 2020.
[234] F. Granata, R. Gargano, G. de Marinis, Artificial intelligence based approaches
to evaluate actual evapotranspiration in wetlands, Sci. Total Environ. 703
(2020) 135653, https://doi.org/10.1016/j.scitotenv.2019.135653.
[235] A. Jahani, B. Rayegani, Forest landscape visual quality evaluation using
artificial intelligence techniques as a decision support system, Stoch. Envi-
ron. Res. Risk Assess. 34 (10) (2020) 1473e1486, https://doi.org/10.1007/
s00477-020-01832-x.
[236] J.A.C. Nunes, I. Cruz, A. Nunes, H.T. Pinheiro, Speeding up coral reef conser-
vation with AI-aided automated image analysis, Nat. Mach. Intell. 2 (6)
(2020), 292-292.
[237] A. Santangeli, Y. Chen, E. Kluen, R. Chirumamilla, J. Tiainen, J. Loehr, Inte-
grating drone-borne thermal imaging with artificial intelligence to locate
bird nests on agricultural land, Sci. Rep. 10 (1) (2020) 10993, https://doi.org/
10.1038/s41598-020-67898-3.
[238] D. Silvestro, S. Goria, T. Sterner, et al., Improving biodiversity protection
through artificial intelligence, Nat. Sustain. 5 (2022) 415e424, https://
doi.org/10.1038/s41893-022-00851-6.
[239] Ivanova, Stoyanka M. Ivanova, Evelina Medarov, Martin, Guidelines for the
application of artificial intelligence in the study of the influence of climate
change on transport infrastructure, in: Conference: VIII International Sci-
entific Conference "Industry 4.0", Summer Session, 28 June - 01 July 2023,
vol. 1, 2023, pp. 29e33. At: Varna.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
29
[240] S. Khalilpourazari, S.H.R. Pasandideh, Designing emergency flood evacuation
plans using robust optimization and artificial intelligence, J. Combin. Optim.
41 (3) (2021) 640e677, https://doi.org/10.1007/s10878-021-00699-0.
[241] D. Rolnick, P.L. Donti, L.H. Kaack, K. Kochanski, A. Lacoste, K. Sankaran,
A.S. Ross, N. Milojevic-Dupont, N. Jaques, A. Waldman-Brown, Tackling
climate change with machine learning, ACM Comput. Surv. 55 (2) (2022)
1e96.
[242] A.L. Stein, Artificial intelligence and climate change, Yale J. Regul. 37 (2020)
890.
[243] A. Buis, Study confirms climate models are getting future warming pro-
jections right, 2020, https://climate.nasa.gov/news/2943/study-confirms-
climate-models-are-getting-future-warming-projections-right/Google.
Scholar, 2020. (Accessed 20 January 2020).
[244] T. Yigitcanlar, K.C. Desouza, L. Butler, F. Roozkhosh, Contributions and risks
of artificial intelligence (AI) in building smarter cities: insights from a sys-
tematic review of the literature, Energies 13 (6) (2020) 1473.
[245] M. Gueriau, F. Cugurullo, R. A. Acheampong and I. Dusparic, "Shared
Autonomous Mobility on Demand: A Learning-Based Approach and Its
Performance in the Presence of Traffic Congestion," in IEEE Intelligent
Transportation Systems Magazine, vol. 12, no. 4, pp. 208-218, winter 2020,
doi: 10.1109/MITS.2020.3014417.J.
[246] F. Cugurullo, Urban artificial intelligence: from automation to autonomy in
the smart city, Front. Sustain. Cities 2 (2020) 38, https://doi.org/10.3389/
frsc.2020.00038.
[247] R.A. Acheampong, F. Cugurullo, Capturing the behavioural determinants
behind the adoption of autonomous vehicles: conceptual frameworks and
measurement models to predict public transport, sharing and ownership
trends of self-driving cars, Transport. Res. F Traffic Psychol. Behav. 62 (2019)
349e375.
[248] F. Cugurullo, R.A. Acheampong, M. Gueriau, I. Dusparic, The transition to
autonomous cars, the redesign of cities and the future of urban sustain-
ability, Urban Geogr. 42 (6) (2021) 833e859.
[249] J. Hawkins, K. Nurul Habib, Integrated models of land use and transportation
for the autonomous vehicle revolution, Transport Rev. 39 (1) (2019) 66e83,
https://doi.org/10.1080/01441647.2018.1449033.
[250] T. Han, K. Muhammad, T. Hussain, et al., An efficient deep learning frame-
work for intelligent energy management in IoT networks, IEEE IoT J 8 (2020)
3170e3179.
[251] S. El Himer, M. Ouaissa, M. Ouaissa, Z. Boulouard, Artificial intelligence of
things (AIoT) for renewable energies systems, in: S. El Himer, M. Ouaissa,
A.A.A. Emhemed, M. Ouaissa, Z. Boulouard (Eds.), Artificial Intelligence of
Things for Smart Green Energy Management. Studies in Systems, Decision
and Control, vol. 446, Springer, Cham, 2022, https://doi.org/10.1007/978-3-
031-04851-7_1.
[252] A. Sleem, I. Elhenawy, Survey of artificial intelligence of things for smart
buildings: a closer outlook, J. Intell. Syst. Internet Things 8 (2) (2023) 63e71,
https://doi.org/10.54216/JISIoT.080206.
[253] Q. Qiao, F. Tao, H. Wu, X. Yu, M. Zhang, Optimization of a capacitated vehicle
routing problem for sustainable municipal solid waste collection manage-
ment using the PSO-TS algorithm, Int. J. Environ. Res. Publ. Health 17 (6)
(2020) 2163.
[254] H.L. Vu, D. Bolingbroke, K.T.W. Ng, B. Fallah, Assessment of waste charac-
teristics and their impact on GIS vehicle collection route optimization using
ANN waste forecasts, Waste Manag. 88 (2019) 118e130.
[255] M. Akhtar, M.A. Hannan, R.A. Begum, H. Basri, E. Scavino, Backtracking search
algorithm in CVRP models for efficient solid waste collection and route
optimization, Waste Manag. 61 (2017) 117e128.
[256] M. Mi, A. Dordevic, A.K. Arsi
c, The optimization of vehicle routing of
communal waste in an urban environment using a nearest neighbirs' algo-
rithm and genetic algorithm: communal waste vehicle routing optimization
in urban areas, in: 2017 Ninth International Conference on Advanced
Computational Intelligence (ICACI), IEEE, 2017, pp. 264e271.
[257] M.A. Hannan, R.A. Begum, A.Q. Al-Shetwi, P.J. Ker, M.A. Al Mamun,
A. Hussain, H. Basri, T.M.I. Mahlia, Waste collection route optimisation model
for linking cost saving and emission reduction to achieve sustainable
development goals, Sustain. Cities Soc. 62 (2020) 102393, https://doi.org/
10.1016/j.scs.2020.102393.
[258] G. Lampropoulos, K. Siakas, J. Viana, O. Reinhold, Artificial intelligence,
blockchain, big data analytics, machine learning and data mining in tradi-
tional CRM and Social CRM: a critical review, in: The 21st IEEE/WIC/ACM
International Joint Conference on Web Intelligence and Intelligent Agent
Technology (WI-IAT), 2022, pp. 504e510, https://doi.org/10.1109/WI-
IAT55865.2022.00080. Niagara Falls, Canada.
[259] A. Parmentola, A. Petrillo, I. Tutore, F. De Felice, Is blockchain able to enhance
environmental sustainability? A systematic review and research agenda
from the perspective of Sustainable Development Goals (SDGs), Bus. Strat.
Environ. 31 (1) (2022) 194e217.
[260] Z. Zheng, S. Xie, H.N. Dai, X. Chen, H. Wang, Blockchain challenges and op-
portunities: a survey, Int. J. Web Grid Serv. 14 (4) (2018) 352, https://doi.org/
10.1504/ijwgs.2018.095647.
[261] D. Siakas, G. Lampropoulos, H. Rahanu, E. Georgiadou, K. Siakas, Emerging
technologies enabling the transition toward a sustainable and circular
economy: the 4R sustainability framework, in: M. Yilmaz, P. Clarke, A. Riel,
R. Messnarz (Eds.), Systems, Software and Services Process Improvement.
EuroSPI 2023. Communications in Computer and Information Science, vol.
1891, Springer, Cham, 2023, https://doi.org/10.1007/978-3-031-42310-9_12.
[262] S.-K. Kim, J.-H. Huh, Blockchain of carbon trading for UN sustainable
development goals, Sustainability 12 (10) (2020) 4021, https://doi.org/
10.3390/su12104021.
[263] Y. Miao, M. Zhou, A. Ghoneim, Blockchain and AI-based natural gas industrial
IoT system: architecture and design issues, IEEE Netw 34 (5) (2020) 84e90.
[264] M. Kouhizadeh, J. Sarkis, Blockchain practices, potentials, and perspectives in
greening supply chains, Sustainability 10 (10) (2018) 3652.
[265] S. Wang, X. Qu, Blockchain applications in shipping, transportation, logistics,
and supply chain, in: Smart Transportation Systems 2019, Springer,
Singapore, 2019, pp. 225e231.
[266] K. Sankaran, Carbon emission and plastic pollution: how circular economy,
blockchain, and artificial intelligence support energy transition? J. Innovat.
Manage. 7 (4) (2019) 7e13.
[267] K. Xiao, W. Shi, Z. Gao, et al., DAER: a resource preallocation algorithm of
edge computing server by using blockchain in intelligent driving, IEEE IoT J 7
(10) (2020) 9291e9302.
[268] R. Liaqat, I.A. Sajjad, M. Waseem, H.H. Alhelou, Appliance level energy
characterization of residential electricity demand: prospects, challenges and
recommendations, IEEE Access 9 (2021) 148676e148697, https://doi.org/
10.1109/ACCESS.2021.3123196.
[269] H. Mora, J.C. Mendoza-Tello, E.G. Varela-Guzm
an, J. Szymanski, Blockchain
technologies to address smart city and society challenges, Comput. Hum.
Behav. 122 (2021) 106854, https://doi.org/10.1016/j.chb.2021.106854.
[270] S. Makani, R. Pittala, E. Alsayed, et al., A survey of blockchain applications in
sustainable and smart cities, Cluster Comput. 25 (2022) 3915e3936, https://
doi.org/10.1007/s10586-022-03625-z.
[271] R. Vinuesa, H. Azizpour, I. Leite, M. Balaam, V. Dignum, S. Domisch, et al., The
role of artificial intelligence in achieving the Sustainable Development Goals,
Nat. Commun. 11 (1) (2020) 233, https://doi.org/10.1038/s41467-019-
14108-y.
[272] S. Blasi, A. Ganzaroli, I. De Noni, Smartening sustainable development in
cities: strengthening the theoretical linkage between smart cities and SDGs,
Sustain. Cities Soc. 80 (2022) 103793, https://doi.org/10.1016/
j.scs.2022.103793.
[273] A. Zaidi, S.-S.M. Ajibade, M. Musa, F.V. Bekun, New insights into the research
landscape on the application of artificial intelligence in sustainable smart
cities: a bibliometric mapping and network analysis approach, Int. J. Energy
Econ. Pol. 13 (4) (2023) 287e299, https://doi.org/10.32479/ijeep.14683.
[274] Y. Duan, J.S. Edwards, Y.K. Dwivedi, Artificial intelligence for decision making
in the era of Big Data eevolution, challenges and research agenda, Int. J. Inf.
Manag. 48 (2019) 63e71.
[275] A.C. S¸ erban, M.D. Lytras, Artificial intelligence for smart renewable energy
sector in Europedsmart energy infrastructures for the next generation
smart cities, IEEE Access 8 (2020) 77364e77737.
[276] L. Bottarelli, M. Bicego, J. Blum, A. Farinelli, Orienteering-based informative
path planning for environmental monitoring, Eng. Appl. Artif. Intell. 77
(2019) 46e58, https://doi.org/10.1016/j.engappai.2018.09.015.
[277] J. Lu, L. Feng, J. Yang, M.M. Hassan, A. Alelaiwi, I. Humar, Artificial agent: the
fusion of artificial intelligence and a mobile agent for energy-efficient traffic
control in wireless sensor networks, Future Generat. Comput. Syst. 95 (2019)
45e51, https://doi.org/10.1016/j.future.2018.12.024.
[278] T. Walsh, A. Evatt, C.S. de Witt, Artificial Intelligence &Climate Change:
Supplementary Impact Report. Oxford, vol. 1, 2020, pp. 1e15.
[279] S.E. Bibri, Transitioning from smart cities to smarter cities: the future po-
tential of ICT of pervasive computing for advancing environmental sustain-
ability, in: Smart Sustainable Cities of the Future, Springer, 2018,
pp. 535e599.
[280] B. Brevini, Black boxes, not green: mythologizing artificial intelligence and
omitting the environment, Big Data Soc. 7 (2) (2020) 2053951720935141,
https://doi.org/10.1177/2053951720935141.
[281] P. Dauvergne, The globalization of artificial intelligence: consequences for
the politics of environmentalism, Globalizations 18 (2) (2021) 285e299,
https://doi.org/10.1080/14747731.2020.1785670.
[282] F.A. Almalki, S.H. Alsamhi, R. Sahal, J. Hassan, A. Hawbani, N.S. Rajput, A. Saif,
J. Morgan, J. Breslin, Green IoT for eco-friendly and sustainable smart cities:
future directions and opportunities, Mobile Network. Appl. (2021), https://
doi.org/10.1007/s11036-021-01790-w.
[283] Z. Wang, H.-T. Liao, Towards the eco-design of Artificial Intelligence and Big
Data applications: a bibliometric analysis of related research, IOP Conf. Ser.
Mater. Sci. Eng. 806 (1) (2020) 5. Article ID 012039.
[284] M.D. Belkin, S. Hsu, S. Ma, Mandal Reconciling Modern Machine-Learning
Practice and the Classical BiaseVariance Trade-Off Proceedings of the Na-
tional Academy of Sciences of the United States of America, vol. 116, 2019,
pp. 15849e15854, 32.
[285] I. Ahmed, G. Jeon, F. Piccialli, From artificial intelligence to eXplainable
artificial intelligence in industry 4.0: a survey on what, how and where, IEEE
Trans. Ind. Inf. 18 (2022) 5031e5042.
[286] I. García-Magari~
no, R. Muttukrishnan, J. Lloret, Human-centric AI for trust-
worthy IoT systems with explainable multilayer perceptrons, IEEE Access 7
(2019) 125562e125574.
[287] Advances in explainable AI applications for smart cities, in: M.M. Ghonge,
N. Pradeep, N. Jhanjhi, P. Kulkarni (Eds.), IGI Global (2023), https://doi.org/
10.4018/978-1-6684-6361-1.
[288] A.R. Javed, W. Ahmed, S. Pandya, P.K.R. Maddikunta, M. Alazab,
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
30
T.R. Gadekallu, A survey of explainable artificial intelligence for smart cities,
Electronics 12 (4) (2023) 1020, https://doi.org/10.3390/electronics12041020.
[289] P. Linardatos, V. Papastefanopoulos, S. Kotsiantis, A.I. Explainable, A review
of machine learning interpretability methods, Entropy 23 (2021) 18.
[290] M. Mayuri, P. Vasile, C. Indranath, Explainable AI: Foundations, Methodolo-
gies and Applications, Springer, 2023, https://doi.org/10.1007/978-3-031-
12807-3.
[291] D. Thakker, B.K. Mishra, A. Abdullatif, S. Mazumdar, S. Simpson, Explainable
artificial intelligence for developing smart cities solutions, Smart Cities 3
(2020) 1353e1382.
[292] Kabir, M.H.; Hasan, K.F.; Hasan, M.K.; Ansari, K. Explainable Artificial Intel-
ligence for Smart City Application: A Secure and Trusted Platform. arXiv
2021, arXiv:2111.00601.
[293] A.L. Hoffmann, Where fairness fails: data, algorithms, and the limits of
antidiscrimination discourse, Inf. Commun. Soc. 22 (7) (2019) 900e915,
https://doi.org/10.1080/1369118X.2019.1573912.
[294] C. O'Neil, Weapons of Math Destruction: How Big Data Increase Inequality
and Threatens Democracy, Penguin, 2016.
[295] S. Larsson, F. Heintz, Transparency in artificial intelligence, Internet Pol. Rev
9 (2020) 1e12.
[296] M.A. Ahmad, A. Teredesai, C. Eckert, Fairness, accountability, transparency in
AI at scale: lessons from national programs, in: Proceedings of the 2020
Conference on Fairness, Accountability, and Transparency, Barcelona, Spain,
27e30 January 2020, pp. 690e699.
[297] S. Burton, I. Habli, T. Lawton, J. McDermid, P. Morgan, Z. Porter, Mind the
gaps: assuring the safety of autonomous systems from an engineering,
ethical, and legal perspective, Artif. Intell. 279 (2020) 103201, https://
doi.org/10.1016/j.artint.2019.103201.
[298] A. Taeihagh, H.S. Lim, Governing autonomous vehicles: emerging responses
for safety, liability, privacy, cybersecurity, and industry risks, Transport Rev.
39 (2019) 103e128. $.
[299] L. Floridi, J. Cowls, T.C. King, M. Taddeo, How to design AI for social good:
seven Essential factors, Sci. Eng. Ethics 26 (2020) 1771e1796.
[300] M. Petit, Toward a critique of algorithmic reason. A state-of-the-art review of
artificial intelligence, its influence on politics and its regulation, Quaderns
del CAC 44 (2018).
[301] A. Matthias, The responsibility gap: ascribing responsibility for the actions of
learning automata, Ethics Inf. Technol. 6 (3) (2004) 175e183, https://doi.org/
10.1007/s10676-004-3422-1.
[302] A. Barredo Arrieta, N. Díaz-Rodríguez, J. Del Ser, A. Bennetot, S. Tabik,
A. Barbado, S. Garcia, S. Gil-Lopez, D. Molina, R. Benjamins, R. Chatila,
F. Herrera, Explainable Artificial Intelligence (XAI): concepts, taxonomies,
opportunities and challenges toward responsible AI, Inf. Fusion 58 (2020)
82e115. https://doi.org/10.1016/j.inffus.2019.12.012.
[303] E. Kassens-Noor, A. Hintze, Cities of the future? The potential impact of
artificial intelligence, AI 1 (2020) 192e197, https://doi.org/10.3390/
ai1020012.
[304] S. Joss, R. Cowley, National policies for local urban sustainability: a new
governance approach? in: M. Eames, T. Dixon, M. Hunt, S. Lannon (Eds.),
Retrofitting Cities for Tomorrow’s World Wiley-Blackwell, Oxford, 2017.
[305] W. Sun, P. Bocchini, B.D. Davison, Applications of artificial intelligence for
disaster management, Nat Hazards 103 (2020) 2631e2689, https://doi.org/
10.1007/s11069-020-04124-3.
S.E. Bibri, J. Krogstie, A. Kaboli et al. Environmental Science and Ecotechnology 19 (2024) 100330
31
