Artificial Intelligence in Sustainable Smart Cities: A Systematic

Study on Applications, Benefits, Challenges, and Solutions

Simone C. dos Santos

, Jéssyka F. F. Vilela

, Thiago H. Carvalho, Thiago C. Rocha,

Thales B. Candido, Vinícius S. Bezerra and Daniel J. Silva

Centro de Informática, Federal University of Pernambuco, Recife, Brazil

Keywords: Smart Cities, Sustainability, Artificial Intelligence, Systematic Literature Review.

Abstract: In an era marked by rapid urban growth and environmental challenges, the advent of “smart cities” holds

promise for a sustainable future. Central to the operational efficiency of these cities is the role of Artificial

Intelligence (AI). This Systematic Literature Review addresses the critical question: How can AI be used in

sustainable smart cities? Using Kitchenham’s guidelines, the review followed a three-step Planning,

Conducting, and Reporting process. Through a comprehensive search in the databases ACM, IEEEXplore,

Scopus, Science Direct, and Emerald, a total of 46 high-quality papers were identified. These papers were

scrutinized to understand the AI services utilized in smart cities, the benefits, the challenges of implementation,

and potential solutions to these challenges. Findings reveal that AI’s impact is multi-dimensional, affecting

transportation, energy management, and citizen engagement, among other areas. However, several challenges

remain, considering ethics and data management. This review serves as an exhaustive guide for researchers

and policymakers interested in leveraging AI for sustainable urban development.

1 INTRODUCTION

In an age of rapid urbanization and environmental

challenges, a ”smart city” has emerged as a beacon of

hope for a more sustainable and efficient future. A

smart city is constantly evolving and, therefore,

requires constant communication and dissemination

of information (Zubizarreta et al., 2016). At the heart

of these smart cities lies the transformative power of

Artificial Intelligence (AI), which can be defined as

an information processing system capable of

generating new non-trivial information processing

systems (Suleimenov et al., 2020). With this in mind,

AI emerges as a vital ally,- enabling municipalities to

harness data-driven insights and make informed

decisions that can lead to creating genuinely

sustainable smart cities. In line with the original

definition of sustainable development, a city can be

defined as sustainable “if its conditions of production

do not destroy over time the conditions of its

reproduction” (Imperatives, 1987) (Castells,

2000)(Ahvenniemi et al., 2017). The fusion of AI

https://orcid.org/0000-0002-7903-9981

https://orcid.org/0000-0002-5541-5188

technology and urban planning represents a paradigm

shift in how cities operate, manage resources, and

serve their inhabitants; intelligent machines are

utilized for making smart decisions and for removing

human tasks in various fields like automatic sensing

applications, medical applications, automated

farming, and automated vehicle driving (Wang et al.,

2019)(Sharma et al., 2021). It goes beyond mere

automation; it’s about creating cities that are not just

smart but also sustainable, livable, and resilient.

This article delves into the pivotal role of AI in

shaping the future of our urban landscapes, exploring

how it is helping to build greener, more efficient, and

socially inclusive cities. From optimizing

transportation systems to managing energy

consumption and fostering citizen engagement, AI is

driving a revolution that promises to redefine urban

living for future generations. However, there is a need

to carry out a systematic study that answers the

following central research question: RQ) How can AI

be used in sustainable smart cities, considering AI-

based services, its benefits, challenges, and solutions

for these challenges?

644

Santos, S., Vilela, J., Carvalho, T., Rocha, T., Candido, T., Bezerra, V. and Silva, D.

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions.

DOI: 10.5220/0012617900003690

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 26th International Conference on Enterprise Information Systems (ICEIS 2024) - Volume 1, pages 644-655

ISBN: 978-989-758-692-7; ISSN: 2184-4992

This paper is structured in five sections to discuss

research results. After this brief introduction, we

describe primary concepts and related work in

Section 2. Section 3 details the research method based

on the Systematic Literature Review. Section 4

presents the results and discusses our research

secondary questions. Finally, Section 5 presents

conclusions and future work.

2 BACKGROUND

2.1 AI Mechanisms and Technologies

Across Industries

AI-based services harness various mechanisms and

technologies, revolutionizing multiple industries (Lee

Park, 2019; Li et al., 2021; Lee Yoon, 2021).

Machine Learning (ML) stands out as a

cornerstone, enabling systems to learn from data,

adapt, and improve performance over time (Singh,

2007). Supervised Learning trains models on labelled

datasets, while unsupervised Learning discovers

patterns in unlabelled data. Reinforcement learning

introduces decision-making through reward-based

systems.

Deep Learning, a subset of ML, utilizes neural

networks with multiple layers to process complex

data representations, excelling in tasks such as image

and speech recognition (Mathew et al., 2021).

Convolutional Neural Networks (CNNs) are pivotal

for visual data (Li et al., 2021), while Recurrent

Neural Networks (RNNs) excel in sequential data like

language (Medsker, 2001).

Natural Language Processing (NLP) empowers

AI to comprehend and generate human language.

This technology underpins virtual assistants, chat-

bots, and language translation services, facilitating

seamless human-computer interaction. Computer

Vision extends AI capabilities to interpret and

understand visual information (Kothadiya et al.,

2021). Image recognition, object detection, and facial

recognition are prominent applications that enhance

the healthcare, security, and retail sectors.

Other AI mechanisms include expert systems,

rule-based systems that mimic human expertise, and

reinforcement learning for autonomous decision-

making (Shrestha et al., 2019). Speech recognition

technology enables voice-based interactions, while

robotics integrates AI for physical tasks in industries

like manufacturing and healthcare. Overall, the

synergy of these mechanisms fuels AI-based services,

transforming industries by automating processes,

improving user experiences, and fostering innovative

solutions across diverse domains (Noor et al., 2022).

2.2 Related Works

This section summarizes ten systematic studies of the

literature on Sustainable Smart Cities, covering

various topics from energy management and

transportation to governance and data security.

Alotaibi et al. (2020) examine recent literature on

smart grids, highlighting their role in enhancing

energy efficiency and sustainability in urban

environments. Also concerning energy efficiency,

Khaje Nasiri et al. (2017) investigate energy-efficient

IoT applications in sustainable solutions,

emphasizing reducing environmental impact in smart

urban environments.

With a focus on IoT applications, Vaidian et al.

(2019) explore their impact on urban mobility,

providing insights into sustainable transportation

solutions. Concerning smart buildings, Ejidike

Mewomo (2023) explores technologies and strategies

for sustainable construction and operation,

contributing to eco-friendly urban development. Also

relating to human well-being, Nitoslawski et al.

(2019) highlight their importance in promoting

environmental sustainability and residents’ well-

being through green spaces in smart cities.

Regarding smart city management, Abdullah et al.

(2019) examine waste management literature,

focusing on smart solutions, such as IoT-enabled

optimization, for efficient and sustainable urban

waste management. Analyzing governance literature,

Pereira et al. (2018) explore models and frameworks

for effective smart city governance, contributing to

sustainable urban development.

Concerning technological concerns, Wong et al.

(2022) present a review investigating blockchain’s

potential and applications for sustainable urban

development, emphasizing transparency and

decentralized systems. Shehab et al. (2021) discuss

connectivity, emphasizing high-speed, low-latency

communication for improved urban services and

sustainability. Finally, Ismagilova et al. (2020) assess

the literature on data security and privacy in smart

cities, addressing concerns and proposing strategies

to ensure secure and responsible data management.

Considering the various aspects and themes

discussed in each of these literature reviews, the

current study proposes a consolidated perspective

from a holistic view of how AI can be used in smart

cities, its benefits, challenges, and potential solutions

for these challenges. A summary of related works can

be found in Table 1.

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

645

Table 1: Related works.

3 METHOD

A systematic literature review (SLR) was carried out

to collect data. This research has followed

Kitchenham et al.'s (2009) guidelines for conducting

a Systematic Literature Review. They defined an SLR

as “a means of identifying, evaluating, and

interpreting all available research relevant to a

particular research question, topic area, or

phenomenon of interest.” This is the primary rationale

for doing an SLR because if research isn’t thorough

on the existing literature, it is of little scientific value.

Using this method, it is possible to get a structured

view of how smart cities have been implementing AI

sustainably by reviewing studies written about topics

related to the area. Also, a well-defined methodology

makes it less likely that the results are biased, gives a

wide range of information across different settings

and empirical methods, and provides the possibility

of turning it into real data visualization (Kitchenham

et al., 2009).

This review was divided into three stages:

Planning, Conducting, and Reporting.

3.1 Planning the Review

Developing research questions that could guide the

search and selection processes was necessary to plan

out the SLR successfully. So, we unfold the central

research question (How can AI be used in sustainable

smart cities?) in the following secondary questions:

 RQ1: What AI services are used in smart cities

for sustainability?

 RQ2: What are the benefits related to the

implementation of AI in the context of RQ1?

 RQ3: What are the challenges associated with

implementing AI in the context of RQ1?

 RQ4: What solutions to challenges

encountered?

The aim of the first research question (RQ1) is to

establish a foundational understanding of the specific

AI applications currently employed in smart cities

that revolve around sustainability. The second

research question (RQ2) aims to uncover the positive

outcomes and advantages of incorporating AI in the

pursuit of sustainability in smart urban environments.

The third research question (RQ3) focuses on

identifying and comprehensively assessing the

hurdles and complexities that arise when

implementing AI into the sustainability framework of

smart cities. Finally, the fourth research question

(RQ4) explores and presents potential remedies and

strategies to overcome the challenges of integrating

AI for sustainability in smart cities.

A generic search string was constructed to explore

these research questions. It was developed to

encapsulate key concepts, ensuring a focused and

comprehensive retrieval of relevant literature.

Specifically, it encompasses the following keywords:

artificial intelligence, sustainable, and smart cities.

Thus, the final search string developed was as

follows:

(“Artificial Intelligence” OR AI OR “Machine

Learning” OR “Deep Learning”) AND

(sustainability OR sustainable OR green) AND

(“smart cities”).

Ref Ti t l e Summar y

(Alotaibi et al.,

2020)

A comprehensive review of recent

advances in smart grids: A

sustainable future with renewable

energy resources.

Focuses on the role of smart

grids in enhancing energy

efficiency in urban areas.

(Vaidian et al.,

2019)

Impact of Internet of Things on

Urban Mobility.

Explores the impact of IoT

applications on sustainable

transportation solutions in

urban settings.

(Wong et al.,

2022)

Potential integration of blockchain

technology into the smart

sustainable city (SSC)

developments: a systematic

review.

Investigates the potential

applications of blockchain in

promoting sustainability and

decentralized systems in urban

development.

(Shehab et al.,

2021)

5G networks towards smart and

sustainable cities: A review of

recent developments, applications

and future perspectives.

Examines the literature on how

5G technology contributes to

improved urban services and

sustainability.

(Ejidike and

Mewomo, 2023)

Benefits of adopting smart

building technologies in building

construction of developing

countries: Review of the literature.

Analyses literature on

technologies and strategies for

sustainable construction and

operation of smart buildings.

(Abdullah et al.,

2019)

IoT-based smart waste

management system in a smart

city.

Focuses on IoT-enabled

optimization and other smart

solutions for efficient and

sustainable urban waste

management.

(Nitoslawski et

al., 2019)

Smarter ecosystems for smarter

cities? A review of trends,

technologies, and turning points

for smart urban forestry.

Reviews literature on energy-

efficient IoT ap- plications,

emphasizing reducing

environmental impact in smart

urban environments.

(Khajenasiri et al.,

2017)

A review on Internet of Things

solutions for intelligent energy

control in buildings for smart city

applications.

Analyses governance literature,

exploring models and

frameworks for effective smart

city governance to contribute

to sustainable urban

development.

(Pereira et al.,

2018)

Smart governance in the context of

smart cities: A literature review.

Addresses concerns and

proposes data security and

privacy strategies in smart

cities, ensuring secure and

responsible data management.

(Ismagilova et al.,

2020)

Security, privacy and risks within

smart cities: Literature review and

development of a smart city

interaction framework.

Discusses several aspects of

how AI can be used in smart

cities, its benefits, challenges,

and potential solutions for

these challenges.

Carvalho et al. Current study. Consolidated perspective from

a holistic view of how AI can

be used in smart cities.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

646

3.2 Conducting the Review

The research methodology employed for this study

involved an automated search conducted across five

prominent research databases: ACM, IEEEXplore,

Scopus, Science Direct, and Emerald. These

platforms were selected for their significant

representation within the computing domain. Since

the topic of implementing AI in smart cities is of

considerable recency (Herath and Mittal, 2022), the

search scope was restricted to papers published in the

last five years, considering the period from 2018 to

2023. Besides, this period was chosen, considering

the analysis of this topic from a trend perspective. In

the initial search, a total of 590 results were obtained

across the five databases: ACM (113), IEEEXplore

(79), Scopus (246), Science Direct (111), and

Emerald (41).

Given the substantial volume of retrieved studies,

a three-step selection process was implemented. First,

a pre-selection step applied exclusion criteria based

on the paper’s abstract alone. The exclusion criteria

encompassed studies unavailable for viewing, studies

with paid content, duplicates, those outside the

defined research area, early access studies, and

studies with four pages or fewer. Subsequently, the

second step applied inclusion criteria, which were (1)

studies between 2018 and 2023, (2) studies in

English, (3) studies in the field of computer science

or related areas, and (4) studies that were articles,

conferences, or journals.

After applying all criteria by filtering out within

the title and abstract, a total of 110 studies were left,

and an additional 18 more were approved after

analysing their introduction and conclusion, leaving

with a total of 128 papers.

In the final step, quality criteria were applied,

evaluating aspects such as the adequacy of contextual

description, clarity of study aims, appropriateness of

research methodology, transparency of findings,

relevant and consistent discussions, and that it

answers at least one research question. These criteria

were applied by awarding a grade (0, 0.5, or 1) for

each criterion, totalling a maximum of 5 points per

study.

Therefore, studies that had a score lower than

three were discarded. Furthermore, the study was

excluded if the score for the “Answer at least one

research question” criteria was zero. Nineteen studies

achieved a score of 5 points, five studies scored 4.5

points, twelve studies achieved 4 points, six studies

were awarded 3.5 points, and a further four studies

were given 3 points, thus leaving a total of 46 studies

after all criteria. The selected papers are listed in

Table 2.

Table 2: Selected studies.

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

647

3.3 Limitations and Threats to Validity

Considering the classification of threats to validity

(Wohlin et al., 2012), we observe some threats to

validity.

While employing a Systematic Literature Review

(SLR) offers a structured and comprehensive

approach to synthesizing existing research

(Kitchenham et al., 2009), it is important to recognize

certain inherent limitations and potential threats to the

validity of the findings. One notable consideration is

that the decision to include only English-language

studies may introduce a language bias. Valuable

research in other languages may be overlooked,

potentially excluding important insights from non-

English sources.

Given the review’s focus on studies published

between 2018 and 2023, there is a temporal limitation.

While this timeframe was chosen to encompass the

most up-to-date research, it may inadvertently exclude

older studies with relevant discussions in the field of

AI in sustainable smart cities.

4 METHOD

The studies collection process involved multiple steps

and criteria to ensure quality and relevance. Table 3

provides a detailed view of the process.

Initially, 590 studies were identified from various

databases, including IEEE Xplore, ACM, Scopus,

Science Direct, and Emerald. After applying three

consecutive filters and a quality assessment, the total

number of studies was reduced to 46.

We then analysed the annual trends in article

publication on the subject. Fig. 1 shows the yearly

evolution of the number of studies published on the

subject.

Notably, there was a significant increase in

publications between 2019 and 2022. The year 2019

has three publications, increasing to 8 in 2020, 12 in

2021, and reaching a peak of 16 in 2022. It’s worth

noting that the data for 2023 only includes six studies,

considering that the research occurred in the middle

of this year. This could indicate a continuing upward

interest and research activity trend around the subject.

Following the quantitative reduction in studies,

the quality of these studies was assessed. Table 4

offers insights into the perceived quality of studies by

showing the average scores from various research

databases. Science Direct leads the list with an im-

impressive average of 4.50, closely followed by ACM

with an average of 4.40. Despite having the most

studies at 19, IEEE Xplore has an average score of

4.26. Emerald and Scopus trail with averages of 4.17

and 4.00, respectively. The overall average score

across all databases stands at 4.32.

Table 3: Evolution of the studies collection process.

Figure 1: Yearly evolution of the number of studies.

Table 4: Average Quality Scores per Research Database.

The types of publications in which these studies

appeared also offer insights into the field’s academic

maturity and activity. Fig. 2 shows the distribution of

the 46 selected studies based on the publication type.

Among them, 18 were published in conferences and

28 in journals.

Figure 2: Types of published studies.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

648

The significant number of conference studies

suggests that the study area is active and possibly

emerging, with new findings being frequently

debated. On the other hand, the more significant

presence in journals also suggests academic maturity

of the topic. This scenario indicates a subject with a

good balance, reflecting both innovation and

consolidation in the field of research.

Lastly, Fig. 3 illustrates the distribution of the 46

selected studies by research database. IEEE Xplore

leads the pack with 19 studies, suggesting that this

database might be a primary source for research in

this field. It is followed by Science Direct, which

contributed 14 studies. Scopus and ACM have equal

contributions, each with 5, and Emerald has 3 studies.

Figure 3: Concentration of studies by research database.

4.1 The Potential of AI for Industrial

Applications

The field of sustainable smart cities has witnessed a

burgeoning interest in various AI concepts, each of

which plays a pivotal role in bolstering urban

sustainability and the overall quality of urban life.

This research paper delves into a selection of the most

promising AI concepts within this domain, particularly

emphasizing their applications and implications. In the

context of smart cities striving for sustainability,

diverse artificial intelligence (AI) services emerge as

critical contributors to achieving environmental goals.

As outlined in Fig. 4, these AI services are evaluated

and ranked based on their significance.

At the forefront of this spectrum is Environmental

Monitoring and Management, scoring the highest at

8. AI’s capabilities are harnessed to continuously

monitor and regulate various environmental factors,

such as air quality, water quality, and overall

ecosystem health. By doing so, these AI systems not

only facilitate the identification of pressing

environmental and mental issues but also enable

timely interventions to mitigate and manage these

challenges.

Figure 4: Concentration of benefits related to AI in smart

cities collected in studies.

The second-most important AI service, with a

score of 7, is Energy Management and Efficiency. AI

technologies are indispensable in optimizing energy

consumption and promoting energy-efficient

solutions within smart cities. These systems help

reduce energy consumption and emissions and

encourage the adoption of sustainable energy

technologies, contributing significantly to

environmental sustainability and urban liveability.

Similarly, Transportation and Traffic Management,

with a score of 7, leverages AI to streamline traffic

flow, minimize congestion, and enhance

transportation systems, ultimately reducing emissions

and improving mobility in urban areas.

Data Analysis and Predictive Modelling, rated at

7, enable informed decision-making processes for

sustainability initiatives and urban planning. By

harnessing AI-driven data analysis and predictive

modelling, smart cities can better address

environmental challenges and optimize resource

allocation for maximum impact. In contrast, AI’s role

in Waste Management, Sustainability in Buildings,

Disaster Prevention and Management, Citizen

Engagement and Participation, Urban Planning and

Design, Green Manufacturing and Technology, IoT

for Sustainability, Education and Social Networking,

Smart Hospitality Services, Climate and Weather

Impact Assessment, and Synergies between Techno-

Scientific Domains, all receiving varying scores,

illustrate the diverse applications of AI in promoting

sustainability across different facets of urban life.

These applications collectively contribute to the

development of smart cities that are not only

environmentally sustainable but also energy-efficient

and responsive to the needs of their citizens, all while

minimizing their ecological footprint.

In summary, integrating AI services into smart

cities is a multifaceted and powerful approach to

addressing environmental challenges and enhancing

urban living while prioritizing sustainability. This

research underscores the critical role AI plays in

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

649

shaping the future of cities, where technology and

data-driven solutions work in tandem to create more

resilient and environmentally conscious urban

environments.

4.2 What Are the Benefits Related to

Implementing AI in Smart Cities

for Sustainability?

Integrating Artificial Intelligence (AI) into the

infrastructure of smart cities holds substantial

promise in advancing sustainability across a broad

spectrum of domains. As elucidated in Fig. 5, this

implementation offers multifaceted advantages.

Figure 5: Concentration of benefits related to AI in smart

cities collected in studies.

Firstly, in Environmental Monitoring and

Management, AI can revolutionize real-time

monitoring, enabling cities to vigilantly track

pollution levels, air quality, and water quality,

thereby facilitating the prompt implementation of

remedial measures to mitigate environmental harm.

The application of AI-driven systems in Energy

Management and Efficiency stands to optimize

energy consumption in buildings, street lighting, and

public facilities, manifesting as reduced energy

wastage and a subsequent decrease in greenhouse gas

emissions.

Additionally, the utilization of AI in

Transportation and Traffic Management promises

improved traffic flow, reduced congestion, and

enhanced public transportation systems, ultimately

leading to lower fuel consumption and decreased air

pollution. In Data Analysis and Predictive Modelling,

AI is indispensable for data-driven decision-making,

harnessing its ability to analyse vast datasets to

predict future trends and optimize resource allocation.

These applications collectively contribute to the

overarching goal of sustainability in smart cities.

Moreover, AI’s role in Waste Management

streamlines waste collection routes, curtails overflows,

and augments recycling processes, thereby minimizing

landfill usage and fostering recycling. In terms of

sustainability in buildings, AI-driven building

management systems hold the potential to optimize

energy usage, lighting, and temperature control,

resulting in tangible energy savings and reduced

carbon footprints in urban buildings.

Furthermore, AI can lend its capabilities to

Disaster Prevention and Management by bolstering

early warning systems and enhancing disaster

response, ultimately bolstering resilience and

mitigating the environmental impact of disasters.

AI’s influence extends into fostering Citizen

Engagement and Participation, utilizing AI-powered

platforms to promote citizen involvement in

sustainability initiatives, bolstering awareness, and

encouraging responsible behaviours among residents.

In urban planning and design, AI’s role is to

optimize land use and reduce resource consumption,

aiding in designing more sustainable urban layouts.

The convergence of AI and the Internet of Things

(IoT) yields remarkable potential for sustainability,

with real-time monitoring and control of critical

aspects such as water and energy usage. AI-driven

educational platforms and social networks contribute

to sustainability awareness and education among city

residents, while in the domain of Smart Hospitality

Services, AI enhances energy efficiency and resource

management in hotels and hospitality facilities,

aligning with sustainability objectives. Moreover, AI

can assist in Climate and Weather Impact

Assessment, supporting the evaluation of climate

change impacts and predicting extreme weather

events, which aids in proactive preparation and

adaptation.

Lastly, AI catalyses synergies between Techno-

Scientific Domains, facilitating collaboration and

knowledge sharing among diverse scientific

disciplines, leading to innovative and sustainable

solutions. Incorporating AI into smart cities is a

pivotal catalyst in advancing sustainability goals. Its

manifold applications optimize resource usage,

enhance environmental monitoring, and stimulate

citizen engagement, collectively fostering the

development of more efficient, eco-friendly, and

resilient urban environments.

4.3 What Are the Challenges

Associated with Implementing AI

in Smart Cities for Sustainability?

Pursuing sustainability and smart city goals through

integrating AI technologies is undoubtedly a

promising avenue, but it is not devoid of significant

challenges. Fig. 6 summarizes the challenges found.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

650

Figure 6: Concentration of challenges by collected studies.

Firstly, the heightened expectations placed on AI

solutions in the context of smart cities necessitate a

robust urban infrastructure capable of accommodating

and sustaining these technologies (S. Wu et al., 2020).

Although imperative, this infrastructure requirement is

a formidable challenge, often requiring substantial

investments in physical and digital systems.

Implementing and evaluating AI algorithms in the

smart city environment introduces complexity.

Algorithms, particularly those designed for traffic

management, exhibit intricate behaviours that defy

simple assessment. Comparing the performance of

traffic controllers by merely controlling the average

vehicular flow at intersections with and without them

belies the intricacies that lie beneath. While

promising for enhancing urban operations, the

application of reinforcement learning techniques

faces hurdles. The inherent complexity of the

algorithms, combined with the stochastic nature of

factors like quest location generation, impedes the

convergence of reinforcement learning processes.

Consequently, achieving the desired outcomes in AI

applications for smart cities becomes daunting.

AI’s contribution to increased electricity

consumption in terms of computational power and

data transmission has raised legitimate environmental

concerns. The carbon emissions associated with this

heightened energy usage are substantial and merit

careful consideration (Yigitcanlar et al., 2021).

Furthermore, the potential for errors in AI-driven

critical decision-making processes, driven by user

and data bias, poses another formidable challenge that

demands rigorous mitigation measures. Processing

vast volumes of transit data to extract actionable

insights presents a multifaceted challenge (Zhang et

al., 2021). The pre-processing required to cleanse and

transform this data into valuable information can be

resource-intensive and time-consuming, impeding

the timely utilization of insight for urban planning

and management.

Additionally, limitations in data collection can

undermine the relevance and accuracy of research

findings. For instance, the study’s data aggregation

concluded in August 2020, rendering it incapable of

capturing subsequent developments, such as the

COVID-19 outbreak that began in December 2019.

These temporal gaps can compromise the efficacy of

AI models in responding to dynamic urban challenges.

Lastly, the domain of energy demand prediction,

which relies heavily on machine learning and statistical

models, poses challenges. Implementing these models

effectively and ensuring their accuracy and adaptabi-

lity to changing circumstances is a non-trivial endea-

vour that requires ongoing attention and refinement.

In the corpus of studies, particular challenges

emerged as recurrent themes, underscoring their

critical importance. Data Handling and Processing

stood out as a ubiquitous challenge, emphasizing the

need for efficient and streamlined data management

processes. Privacy, Security, and Ethics resurfaced

consistently, highlighting the imperative of

establishing robust frameworks to address these

ethical and security concerns. Furthermore, the

intricate task of seamlessly integrating Infrastructure

and Technology emerged as a persistent challenge,

signifying its central role in successfully adopting AI

in smart cities for sustainability.

4.4 What Solutions to the Challenges

Encountered?

Fig. 7 shows the proposed solutions for integrating

artificial intelligence (AI) into smart cities. These

present a transformative potential, positively

impacting a range of identified problems. The

correspondence between the issues and solutions is

presented in Table 5.

Figure 7: Concentration of solutions by collected studies.

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

651

Table 5: Solutions found.

For example, the approach to enhancing the

interpretability of AI algorithms addresses

algorithmic complexity and tackles concerns about

the transparency and comprehensibility of decision-

making processes. By visualizing crucial regions for

classification, gains are made in performance and

confidence in the results, facilitating the acceptance

and utilization of these advanced technologies.

Another crucial problem is high energy

consumption, a considerable barrier to widespread AI

implementation. The proposal to use more efficient

algorithms has a direct positive impact, reducing the

demand for computational resources and,

consequently, electricity consumption. Moreover, by

reducing energy consumption, efforts are made to

mitigate carbon emissions, aligning with

environmental sustainability objectives.

Pursuing to minimize decision-making errors is

another challenge that the solutions address. By

adopting explainable AI methods, such as SHAP

(SHapley Additive exPlanations), that provide

insights into the factors influencing predictions,

optimizing decisions to improve energy consumption

is possible (Lundberg & Lee, 2017). This more

transparent understanding of decision-making

processes allows for more precise and targeted

adjustments, resulting in more efficient energy

management.

The integration and efficient sharing of data, a

proposed solution for complexity in traffic data

analysis, also resonates with other challenges. In

addition to optimizing traffic management, this

solution provides more precise data on travel patterns,

facilitating the prediction of energy demand. This

accurate prediction is crucial for efficiently allocating

energy resources, enabling a more rational and

sustainable use of the available energy. Regarding the

limitation in data collection related to dynamic

events, the proposal to implement a real-time data

collection system solves this problem, contributing to

a more holistic and real-time view of the urban

environment. This means that the collected data

addresses energy demand and assists in traffic

management and various other areas, enabling more

informed and dynamic urban planning.

Finally, the proposed solutions directly address

the issue of accurately predicting energy demand,

considering its importance for energy efficiency. The

continuous improvement of AI-based energy demand

forecasting models optimizes resource allocation and

contributes to more efficient energy use. The

adaptability of these models to variations in

circumstances and urban energy demands means

more agile and effective energy management, leading

to tangible benefits in terms of efficiency and

sustainability.

In an extensive review of studies conducted,

several recurrent themes emerged as fundamental

solutions to the challenges encountered in integrating

artificial intelligence (AI) into the fabric of smart

cities. Data Collection and Integration, as a

cornerstone, enables a more precise understanding of

urban dynamics. Complementing this, advancements

in AI Algorithms and Technology were highlighted,

paving the way for optimized energy management,

and streamlined traffic flow. Simultaneously,

emphasizing Transparency and Responsibility in AI

systems underscored the necessity of building trust

and comprehension among stakeholders.

A parallel focus on Infrastructure and Technology

Investment signified the essential role of financial

backing for sustainable implementation.

Additionally, ensuring privacy and security in data

handling proved to be a non-negotiable aspect. Lastly,

fostering collaboration through public-private

partnerships and driving awareness via Education and

Awareness initiatives were integral to sustainable

progress. These intertwined solutions collectively

present a roadmap for developing smarter and more

efficient cities.

Thus, the outlined solutions address the

challenges in isolation and interconnect to amplify

positive impacts across a broader spectrum of

problems, contributing to more sustainable and

resource-efficient cities.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

652

5 CONCLUSIONS

This paper has conducted a comprehensive review of

the literature, highlighting the multi-faceted

applications of Artificial Intelligence in the context of

Smart Cities. The synthesis of information presented

here underscores the potential for AI-driven solutions

to tackle complex urban challenges and improve city

dwellers' overall quality of life. By harnessing the

power of AI to analyze and utilize Big Data

effectively, Smart Cities can optimize resource

allocation, enhance infrastructure, and provide

innovative services that cater to the ever-evolving

needs of their populations. Nevertheless, these

advancements must be accompanied by robust ethical

frameworks and stringent data privacy measures to

safeguard individuals' rights and security. The

responsible development and deployment of AI in

Smart Cities should be guided by a commitment to

transparency, fairness, and accountability. In essence,

this paper highlights the promising synergy between

Artificial Intelligence and the burgeoning realm of

Big Data in Smart Cities. As we navigate the path

forward, it is crucial for researchers, policymakers,

and stakeholders to collaboratively address the ethical

and legal challenges while harnessing the full

potential of AI for the benefit of urban communities

worldwide. By doing so, we can realize the vision of

truly smart and sustainable cities that prioritize the

well-being and prosperity of their residents.

Future research could explore AI's ethical, legal,

and data management aspects in Smart Cities,

focusing on privacy, fairness, and accountability.

Collaborative efforts are needed to integrate AI with

Big Data for sustainable urban development, ensuring

technology benefits all residents while responsibly

addressing complex urban challenges and people's

rights.

REFERENCES

Abdullah, N., Alwesabi, O. A., and Abdullah, R. (2019).

Iot-based smart waste management system in a smart

city. In Proceedings of the IRICT 2018, pages 364–371,

Springer International Publishing.

Ahvenniemi, H., Huovila, A., Pinto-Seppa ̈, I., and

Airaksinen, M. (2017). What are the differences

between sustainable and smart cities? Cities, 60:234–

245.

Al Hashlamoun, N., Al Barghuthi, N., and Tamimi, H.

(2023). Exploring the intersection of ai and sustainable

computing: Opportunities, challenges, and a framework

for responsible applications. In 2023 9th International

Conference on Information Technology Trends (ITT),

pages 220–225. IEEE.

Allam, Z. and Dhunny, Z. A. (2019). On big data, artificial

intelligence and smart cities. Cities, 89:80–91.

Alotaibi, I., Abido, M. A., Khalid, M., and V., S. A. (2020).

A comprehensive review of recent advances in smart

grids: A sustainable future with renewable energy re-

sources. Energies, 13(23):6269.

Anedda, M., Fadda, M., Girau, R., Pau, G., and Giusto, D.

(2023). A social smart city for public and private

mobility: A real case study. Computer Networks, 220.

Anthopoulos, L. and Kazantzi, V. (2022). Urban energy

efficiency assessment models from an ai and big data

perspective: Tools for policy makers. Sustainable Cities

and Society, 76:103492.

Buss, E., Rabbel, T.-L., Horvat, V., Krizmancic, M., Bog-

dan, S., Wahby, M., and Hamann, H. (2022). Phyton-

odes for environmental monitoring: Stimulus

classification based on natural plant signals in an

interactive energy-efficient bio-hybrid system. In

Proceedings of the ACM GoodIT ’22, pages 258–264.

Association for Computing Machinery.

Castells, M. (2000). Urban sustainability in the information

age. City, 4(1):118–122.

Ceccarini, C., Delnevo, G., and Prandi, C. (2020). Frugar:

Exploiting deep learning and crowdsourcing for frugal

gardening. In Proceedings of the 1st Workshop on

Experiences with the Design and Implementation of

Frugal Smart Objects, pages 7–11, New York, NY,

USA. Association for Computing Machinery.

Costa, D. G. et al. (2022). A survey of emergencies

management systems in smart cities. IEEE Access,

10:61843– 61872.

DeLong, S. and Tolk, A. (2021). Sustainable computing

and simulation: A literature survey. In 2021 Winter

Simulation Conference (WSC), pages 1–12. IEEE.

Ejidike, C. C. and Mewomo, M. C. (2023). Benefits of

adopting smart building technologies in building con-

struction of developing countries: Review of literature.

SN Applied Sciences, 5(2):52.

Essamlali, I., Bahnasse, A., Khiat, A., and Ouajji, H.

(2022). Machine learning in the service of a clean city.

Procedia Computer Science, 198:530–535.

Ge, L., Sarhani, M., Voß, S., and Xie, L. (2021). Review of

transit data sources: Potentials, challenges and com-

plementarity. Sustainability, 13:11450.

Ghadami, N., Gheibi, M., Kian, Z., Faramarz, M. G.,

Naghedi, R., Eftekhari, M., Fathollahi-Fard, A. M.,

Dulebenets, M. A., and Tian, G. (2021).

Implementation of solar energy in smart cities using an

integration of artificial neural network, photovoltaic

system and classical delphi methods. Sustainable Cities

and Society, 74:103149.

Ghahramani, M., Zhou, M., Molter, A., and Pilla, F. (2022).

Iot-based route recommendation for an intelligent

waste management system. IEEE Internet of Things

Journal, 9(14):11883–11892.

Gohari, A., Ahmad, A. B., Rahim, R. B. A., Supa’at, A. S.

M., Razak, S. A., and Gismalla, M. S. M. (2022).

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

653

Involvement of surveillance drones in smart cities: A

systematic review. IEEE Access, 10:56611–56628.

Gonc ̧alves, D., Sheikhnejad, Y., Oliveira, M., and Mar-

tins, N. (2020). One step forward toward smart city

utopia: Smart building energy management based on

adaptive surrogate modelling. Energy and Buildings,

223:110146.

Grzelczak, M. and Duch, P. (2021). Deep reinforcement

learning algorithms for path planning domain in grid-

like environment. Appl. Sci., 11:11335.

Heidari, A., Navimipour, N. J., and Unal, M. (2022). Ap-

plications of ml/dl in the management of smart cities

and societies based on new trends in information

technologies: A systematic literature review.

Sustainable Cities and Society, 85:104089.

Herath, H. and Mittal, M. (2022). Adoption of artificial

intelligence in smart cities: A comprehensive re- view.

International Journal of Information Manage- ment

Data Insights, 2(1):100076.

Hsu, H. and Tseng, K.-F. (2022). Facing the era of

smartness: constructing a framework of required

technology competencies for hospitality practitioners.

Journal of Hospitality and Tourism Technology,

13(3):500–526.

Imperatives, S. (1987). Report of the world commission on

environment and development: Our common future.

Accessed Feb, 10:1–300.

Ismagilova, E., Hughes, L., Rana, N. P., and Dwivedi, Y.

K. (2020). Security, privacy and risks within smart.

Jafari, S., Shahbazi, Z., and Byun, Y.-C. (2021). Improving

the performance of single-intersection urban traffic

networks based on a model predictive controller.

Sustainability, 13:5630.

Jain, A., Gue, I. H., and Jain, P. (2023). Research trends,

themes, and insights on artificial neural networks for

smart cities towards sdg-11. Journal of Cleaner Pro-

duction, 412.

Jha, A. K., Ghimire, A., Thapa, S., Jha, A. M., and Raj, R.

(2021). A review of ai for urban planning: Towards

building sustainable smart cities. In 2021 6th Inter-

national Conference on ICICT, pages 937–944. IEEE.

Kaya, K., Ak, E., Yaslan, Y., and Oktug, S. F. (2021).

Waste-to-energy framework: An intelligent energy re-

cycling management. Sustainable Computing: Infor-

matics and Systems, 30:100548.

Khajenasiri, I., Estebsari, A., Verhelst, M., and Gielen, G.

(2017). A review on internet of things solutions for

intelligent energy control in buildings for smart city

applications. Energy Procedia, 111:770–779.

Kitchenham, B., Brereton, O. P., Budgen, D., Turner, M.,

Bailey, J., and Linkman, S. (2009). Systematic

literature reviews in software engineering–a systematic

literature review. Information and software technology,

51(1):7–15.

Kothadiya, D., Chaudhari, A., Macwan, R., Patel, K., &

Bhatt, C. (2021). The convergence of deep learning and

computer vision: Smart city applications and research

Challenges. In ICIIC 2021, pp. 14-22. Atlantis Press.

Lee, S. H., Lee, T., Kim, S., and Park, S. (2019). Energy

consumption prediction system based on deep learning

with edge computing. In 2019 IEEE ICET, pages 473–

477, Chengdu, China.

Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., ... & He,

B. (2021). A survey on federated learning systems:

Vision, hype and reality for data privacy and protection.

IEEE Transactions on Knowledge and Data

Engineering, 35(4), 3347-3366.

Liu, H., Li, Y., Fu, Y., Mei, H., Zhou, J., Ma, X., and Xiong,

H. (2020). Polestar: An intelligent, efficient and

national-wide public transportation routing engine. In

Proceedings of the 26th ACM KDD ’20, pages 2321–

2329, New York, NY, USA. Association for

Computing Machinery.

Lundberg, S. M., & Lee, S. I. (2017). A unified approach to

interpreting model predictions. Advances in neural

information processing systems, 30.

Ma, Y., Ping, K., Wu, C., Chen, L., Shi, H., and Chong, D.

(2020). Artificial intelligence powered internet of

things and smart public service. Library Hi Tech,

38(1):165–179.

Mahamuni, C. V., Sayyed, Z., and Mishra, A. (2022).

Machine learning for smart cities: A survey. In 2022

IEEE IPRECON Conference (IPRECON), pp. 1–8.

Mahrez, Z., Sabir, E., Badidi, E., Saad, W., and Sadik, M.

(2022). Smart urban mobility: When mobility systems

meet smart data. IEEE Transactions on Intelligent

Transportation Systems, 23(7):6222–6239.

Mathew, A., Amudha, P., & Sivakumari, S. (2021). Deep

learning techniques: an overview. Advanced Machine

Learning Technologies and Applications: Proceedings

of AMLTA 2020, 599-608.

Medsker, L. R., & Jain, L. (2001). Recurrent neural

networks. Design and Applications, 5(64-67), 2.

Mohanty, P. K., Das, P., and Roy, D. S. (2022a). Predicting

daily household energy usages by using model agnostic

language for exploration and explanation. In

International Conference OCIT, pages 543–547. IEEE.

Mohanty, P. K., Roy, D. S., and Reddy, K. H. K. (2022b).

Explainable ai for predicting daily household energy

usages. In International Conference on Artificial

Intelligence and Data Engineering, pages 182–186.

Mortaheb, R. and Jankowski, P. (2023). Smart city

reimagined: City planning and geoai in the age of big

data. Journal of Urban Management, 12:4–15.

Navarathna, P. J. and Malagi, V. P. (2018). Artificial

intelligence in smart city analysis. In 2018 International

Conference on Smart Systems and Inventive

Technology (ICSSIT), pages 44–47, Tirunelveli, India.

Nitoslawski, S. A., Galle, N. J., Van Den Bosch, C. K., and

Steenberg, J. W. (2019). Smarter ecosystems for

smarter cities? a review of trends, technologies, and

turning points for smart urban forestry. Sustainable

Cities and Society, 51:101770.

Noor, N., Rao Hill, S., & Troshani, I. (2022). Recasting

service quality for AI-based service. Australasian

Marketing Journal, 30(4), 297-312.

Oyinlola, T. (2021). Energy prediction in edge environment

for smart cities. In 2021 IEEE 7th World Forum on

Internet of Things (WF-IoT), pages 439–442. IEEE.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

654

Parlina, A., Ramli, K., and Murfi, H. (2021). Exposing

emerging trends in smart sustainable city research us-

ing deep autoencoders-based fuzzy c-means. Sustain-

ability, 13:2876.

Pereira, G. V., Parycek, P., Falco, E., and Kleinhans, R.

(2018). Smart governance in the context of smart cities:

A literature review. Information Polity, 23(2):143–162.

Piccialli, F., Giampaolo, F., Prezioso, E., Crisci, D., and

Cuomo, S. (2021). Predictive analytics for smart

parking: A deep learning approach in forecasting of iot

data. ACM Transactions on Internet Technology,

21(3):1–21.

Rajkumar, P. V. (2022, December). Gauging Carbon

Footprint of AI/ML Implementations in Smart Cities:

Methods and Challenges. In FMEC (pp. 1-5). IEEE.

Rehena, Z. and Janssen, M. (2018). Towards a framework

for context-aware intelligent traffic management sys-

tem in smart cities. In Proceedings of WWW ’18, pages

893– 898, Republic and Canton of Geneva, CHE.

Ren, Y., Xie, R., Yu, F. R., Huang, T., and Liu, Y. (2022).

Green intelligence networking for connected and

autonomous vehicles in smart cities. IEEE Transactions

on Green Communications and Networking,

6(3):1591–1603.

Schu ̈rholz, D., Kubler, S., and Zaslavsky, A. (2020).

Artificial intelligence-enabled context-aware air quality

prediction for smart cities. Journal of Cleaner

Production, 271:121941.

Selvaraj, R., Kuthadi, V. M., and Baskar, S. (2023). Smart

building energy management and monitoring system

based on artificial intelligence in smart city. Sustainable

Energy Technologies and Assessments, 56.

Shah, D. K., Singh, R., Gehlot, A., Khantwal, S., Ahmad,

A. J., and Akram, S. V. (2022). Smart kitchen: Real

time monitoring of kitchen through iot. In 2022 3rd

International Conference on Intelligent Engineering

and Management (ICIEM), pages 718–722. IEEE.

Sharma, A., Podoplelova, E., Shapovalov, G., Tselykh, A.,

and Tselykh, A. (2021). Sustainable smart cities:

convergence of artificial intelligence and blockchain.

Sustainability, 13(23):13076.

Shehab, M. J., Kassem, I., Kutty, A. A., Kucukvar, M.,

Onat, N., and Khattab, T. (2021). 5g networks towards

smart and sustainable cities: A review of recent

developments, applications and future perspectives.

IEEE Access, 10:2987–3006.

Shrestha, Y. R., Ben-Menahem, S. M., & Von Krogh, G.

(2019). Organizational decision-making structures in

the age of artificial intelligence. California management

review, 61(4), 66-83.

Sirmacek, B. and Vinuesa, R. (2022). Remote sensing and

ai for building climate adaptation applications. Results

in Engineering, 15:100524.

Suleimenov, I. E., Vitulyova, Y. S., Bakirov, A. S., and

Gabrielyan, O. A. (2020). Artificial intelligence: what

is it? In Proceedings of the 2020 6th International

Conference on Computer and Technology Applica-

tions, pages 22–25.

Tanko, B., Essah, E., Elijah, O., Zakka, W., and Klufallah,

M. (2023). Bibliometric analysis, scientometrics and

metasynthesis of internet of things (iot) in smart

buildings. Built Environment Project and Asset Man-

agement, 13(5):646–665.

Vaidian, I., Azmat, M., and Kummer, S. (2019). Impact of

internet of things on urban mobility. In Proceedings of

the Innovation Arabia, pages 44–47.

Wang, K., Dong, J., Wang, Y., and Yin, H. (2019). Securing

data with blockchain and ai. Ieee Access, 7:77981–

77989.

Wohlin, C., Runeson, P., Ho ̈st, M., Ohlsson, M. C., Reg-

nell, B., and Wessle ́n, A. (2012). Experimentation in

software engineering. Springer Science & Business

Media.

Wong, P. F., Chia, F. C., Kiu, M. S., and Lou, E. C. (2022).

Potential integration of blockchain technology into

smart sustainable city (ssc) developments: a systematic

review. Smart and Sustainable Built Environment,

11(3):559–574.

Wu, S. et al. (2020). Evaluation of smart infrastructure

systems and novel uv-oriented solution for integration,

resilience, inclusiveness, and sustainability. In 2020 5th

International Conference on Universal Village (UV),

pages 1–45, Boston, MA, USA.

Yeasmin, S., Syed, S. N. J., Shmais, L. A., and Dubayyan,

R. A. (2020). Artificial intelligence-based co2 emission

predictive analysis system. In 2020 International

Conference on Artificial Intelligence Modern Assistive

Technology (ICAIMAT), pages 1–6, Riyadh, Saudi

Arabia.

Yigitcanlar, T., Mehmood, R., and Corchado, J. (2021).

Green artificial intelligence: Towards an efficient,

sustainable and equitable technology for smart cities

and futures. Sustainability, 13:8952.

Yoon, G. et al. (2019). Prediction of machine learning base

for efficient use of energy infrastructure in smart city.

In International Conference on Computing, Electronics

Communications Engineering, 32–35, London, UK.

Zhang, Y., Geng, P., Sivaparthipan, C., and Muthu, B. A.

(2021). Big data and artificial intelligence based early

risk warning system of fire hazard for smart cities.

Sustainable Energy Technologies and Assessments,

45:100986.

Zubizarreta, I., Seravalli, A., and Arrizabalaga, S. (2016).

Smart city concept: What it is and what it should be.

Journal of Urban Planning and Development,

142(1):04015005.

Artiﬁcial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Beneﬁts, Challenges, and Solutions

655