Article
citation information:
Trivedi,
J., Devi, M.S., Vithalani, C., Parmar, K., Dave, D. Industrial intelligence for smart
cities: the role of AI and IoT in transforming urban mobility and
infrastructure. Scientific Journal of
Silesian University of Technology. Series Transport. 2025, 129, 261-281. ISSN: 0209-3324.
DOI: https://doi.org/10.20858/sjsutst.2025.129.15
Janak TRIVEDI[1],
Mandalapu Sarada DEVI[2],
Chandresh VITHALANI[3],
Kiran PARMAR[4], Dhara DAVE[5]
INDUSTRIAL
INTELLIGENCE FOR SMART CITIES: THE ROLE OF AI AND IOT IN TRANSFORMING URBAN
MOBILITY AND INFRASTRUCTURE
Summary. This review synthesizes
research on AI and IoT in urban mobility, focusing on traffic management,
public transportation systems, and autonomous vehicles to address escalating
urban congestion, environmental impact, and mobility demands. This review aimed
to evaluate AI and IoT applications in traffic flow optimization, benchmark
integration in public transit, identify autonomous vehicle frameworks, compare
predictive models and sensor networks, and analyze
adoption challenges. A systematic analysis of global empirical, simulation, and
theoretical studies was conducted, emphasizing technological convergence,
performance outcomes, data utilization, and barriers. The findings reveal that
AI-driven predictive models combined with IoT sensor networks significantly
improve traffic efficiency and reduce emissions, whereas AI-IoT integration
enhances public transit reliability through predictive maintenance and dynamic
scheduling. Autonomous vehicles, supported by IoT-enabled communication and AI
decision-making, demonstrate the potential for safety and sustainability gains
but face regulatory, infrastructural, and acceptance challenges. Advanced
machine learning techniques optimize real-time data analytics but encounter
scalability and explainability limitations. Collectively, these findings
underscore the transformative potential of AI-IoT in urban mobility, contingent
on addressing privacy, infrastructure, and social factors. The synthesis
highlights the need for interdisciplinary approaches to advance scalable,
secure, and user-centered AI-IoT urban mobility
solutions that inform future research and practical implementations.
Keywords: artificial intelligence, internet of things, intelligent transportation
systems
1.
INTRODUCTION
Urbanization and industrial expansion have driven
the global push toward Smart Cities – digitally enhanced urban spaces that use
real-time data to optimize city services and infrastructure. The Fourth
Industrial Revolution (Industry 4.0) has accelerated this evolution by
embedding Artificial Intelligence (AI) and the Internet of Things (IoT) into
industrial and civic systems. Their convergence, often referred to as
Industrial Intelligence, is central to smart urban environments, particularly
in transportation, infrastructure monitoring, and public service delivery.
This study explores how the integration of AI and
IoT in industrial domains facilitates the transformation of urban mobility and
infrastructure management, thus supporting the Smart City vision. Special
emphasis is placed on Intelligent Transportation Systems (ITS), which are key
to reducing congestion, improving safety, and enhancing the commuter
experience.
Research on AI and IoT in urban mobility has
emerged as a critical area of inquiry owing to the increasing challenges posed
by rapid urbanization, traffic congestion, and environmental concerns in cities
worldwide [2]. Over the past decade, the evolution of ITS has integrated AI,
the Internet of Things, and machine learning to enhance traffic management,
public transportation, and autonomous vehicle operations [4]. These
technologies offer practical solutions for improving urban mobility efficiency,
reducing carbon emissions, and elevating the quality of life of urban residents
[5]. For instance, urban areas face significant economic and environmental
costs owing to congestion, with studies highlighting the urgent need for
adaptive, data-driven traffic control and sustainable transit systems [8].
Despite these advances, persistent problems remain
in the effective management of dynamic traffic flows, optimization of public
transit, and deployment of autonomous vehicles at scale [9]. Existing research
reveals gaps in the real-time integration of heterogeneous data sources,
scalability of AI models in complex urban environments, and ethical
implications of pervasive data collection [13]. Moreover, there is a debate
over the balance between centralized and edge computing approaches for traffic
management and the readiness of the infrastructure to support widespread
autonomous vehicle adoption [16]. Failure to address these gaps risks
exacerbating congestion, safety issues, and environmental degradation,
underscoring the need for comprehensive and adaptive urban mobility frameworks
[18].
The conceptual framework underpinning this review
defines AI as computational methods that enable autonomous decision-making, IoT
as interconnected sensor networks that facilitate real-time data acquisition,
and urban mobility as the movement of people and goods within cities [6]. The
interplay of these concepts forms the basis for intelligent transportation
systems that leverage AI algorithms to analyze
IoT-generated data, thereby enabling predictive traffic management, optimized
public transit, and autonomous vehicle coordination [21]. This framework guides
a systematic examination of how these technologies collectively transform urban
mobility.
This systematic review aims to critically evaluate
the integration of AI and IoT in urban mobility, focusing on traffic
management, public transportation systems, and autonomous vehicles. It aims to
synthesize current knowledge, identify technological and implementation
challenges, and propose directions for future research to bridge the existing
gaps. This review contributes to the literature by providing a holistic
understanding of the state-of-the-art and practical implications of sustainable
urban transportation [23].
The review methodology involved a comprehensive
analysis of recent peer-reviewed studies, emphasizing empirical findings and
theoretical advancements. The inclusion criteria prioritized studies addressing
AI and IoT applications in urban traffic, transit, and autonomous vehicle
contexts. The findings are organized thematically to elucidate technological
trends, challenges, and opportunities, facilitating a structured discourse on
advancing intelligent urban mobility [24].
2. PURPOSE AND SCOPE OF THE
REVIEW
This report aims to examine the existing research
on "AI and IoT in urban mobility, focusing on traffic management, public
transportation systems, and autonomous vehicles" to synthesize current
knowledge, identify technological advancements, and evaluate their impacts on
urban transportation efficiency and sustainability. This review is important
because urban centers face escalating challenges
related to congestion, environmental degradation, and mobility demands. By
critically analyzing how artificial intelligence and
the Internet of Things are integrated into traffic management, public transit,
and autonomous vehicle systems, this report aims to highlight effective
strategies, emerging trends, and gaps in the literature. Ultimately, this study
seeks to inform future research directions and practical implementations that
can foster smarter, safer, and more sustainable urban mobility solutions.
Specific objectives:
· To evaluate the current
knowledge of AI and IoT applications in urban traffic management and congestion
mitigation.
· Benchmarking of existing
approaches to integrating AI and IoT in public transportation systems for
enhanced operational efficiency.
· Identification and synthesis
of technological frameworks enabling autonomous vehicle deployment in smart
city mobility ecosystems.
· To compare the effectiveness
of AI-driven predictive models and IoT-enabled sensor networks in optimizing
urban mobility.
· To deconstruct the challenges
and opportunities related to data privacy, infrastructure, and user acceptance
in AI-IoT urban mobility solutions.
3. LITERATURE SELECTION
This section maps the research landscape of the
literature on AI and IoT in urban mobility, focusing on traffic management,
public transportation systems, and autonomous vehicles, encompassing a broad
spectrum of technological applications and urban contexts. These studies
collectively explore the integration of AI and IoT technologies to enhance
traffic flow, optimize public transit, and facilitate autonomous vehicle
deployment, often emphasizing sustainability and efficiency. Methodologies
range from empirical simulations and system implementations to comprehensive
reviews and theoretical frameworks, with geographic focuses spanning global
smart city initiatives and specific urban case studies.
3.1. Comparative Analysis
This comparative analysis addresses key research
questions by synthesizing technological convergence, performance outcomes,
autonomous vehicle impacts, data utilization, and adoption challenges, thereby
informing future urban mobility innovation (Tab. 1).
Tab. 1
The future of urban mobility innovation
|
Reference Paper |
Technological Integration |
Performance Improvement |
Autonomous Vehicle Impact |
Data Utilization
Efficiency |
Adoption Challenges |
|
[1] |
High AI, IoT, ML fusion in transport systems |
Significant congestion and carbon footprint
reduction |
Limited AV focus, mainly infrastructure support |
Real-time data and predictive maintenance emphasized |
Challenges in dynamic validation and data reliance |
|
[2] |
IoT and AI for real-time urban mobility measurement |
Improved travel experience and infrastructure
monitoring |
AVs as emerging mobility solutions |
Data mining and predictive analytics |
Behavioral
and policy adaptation challenges |
|
[3] |
IoT-enabled ITS with AI for parallel traffic systems |
Simulation-based traffic prediction and control |
AVs included in simulation frameworks |
Big data and iterative simulation for traffic
analysis |
Complexity in system integration and scalability |
|
[4] |
ML applications in next-gen intelligent transport
systems |
Improved safety and energy efficiency |
ML supports cooperative driving and hazard warning |
Large-scale data analytics and ML forecasting |
Scalability and quality-of-service challenges |
|
[5] |
ITS with AI and IoT for sustainable smart cities |
Traffic efficiency and safety improvements |
Includes autonomous vehicle components |
Real-time communication and mobility prediction |
Security and privacy concerns in ITS deployment |
|
[6] |
Integration of IoT, AI, and ML in urban systems |
Enhanced urban efficiency and sustainability |
AVs part of integrated urban mobility solutions |
Predictive analytics and real-time data fusion |
Ethical, privacy, and scalability challenges |
|
[7] |
AI and IoT integrated with ITS for dynamic traffic
control |
Improved traffic flow and congestion mitigation |
Discusses AVs as future direction |
Uses IoT sensors and AI for adaptive management |
Multifaceted barriers including policy and public
engagement |
|
[8] |
AI and IoT for traffic management in large cities |
Improved traffic efficiency and sustainability |
AVs not emphasized |
IoT sensor data combined with AI analytics |
Urban infrastructure and technology adoption |
|
[9] |
AI-driven innovations in intelligent urban transport |
Optimized traffic flow and public transit efficiency |
AV deployment and predictive modelling reviewed |
Data-driven decision-making and AI forecasting |
Data privacy and infrastructure limitations |
|
[10] |
Systematic review of AVs in urban areas |
AVs improve safety, reduce congestion and emissions |
Comprehensive
AV impact assessment |
Data on user attitudes and infrastructure needs |
Regulatory and societal acceptance issues |
|
[11] |
AI perspectives in smart cities for vehicle
automation |
AI enables smart traffic control and vehicle
automation |
Focus on AV automation and driver modelling |
Data annotation and model accuracy challenges |
Trust and explain ability issues in AI systems |
|
[12] |
Advances in AI for transportation system development |
Improved traffic flow and incident management |
Emerging AV technologies and autonomous navigation |
Real-time monitoring and AI-based route optimization |
Computational complexity and privacy concerns |
|
[13] |
IoT and AI integration in intelligent transport
systems |
Increased traffic throughput and reduced emissions |
AVs included in ITS frameworks |
Big data analytics and edge computing |
Cybersecurity and data integrity challenges |
|
[14] |
Edge computing with AI-IoT for energy-efficient
transport |
Reduced emissions and improved freight movement |
AVs indirectly supported via traffic optimization |
Distributed data processing at edge nodes |
Deployment costs and multi-agent coordination |
|
[15] |
ITS using vehicular networks and IoV
with ML models |
High detection accuracy and computing efficiency |
AVs integrated via vehicular networks |
Ensemble learning and feature selection |
Computational efficiency and model scalability |
|
[16] |
IoT and AI-enabled secure AVs for smart cities |
AVs improve traffic and environmental management |
Focus on AV security and communication |
Vehicle-to-vehicle and infrastructure data routing |
Privacy, cybersecurity, and governance challenges |
|
[17] |
AI-driven traffic flow management in smart cities |
Improved transportation efficiency and congestion
reduction |
AVs indirectly supported |
AI traffic prediction and dynamic routing |
Interoperability and ethical considerations |
|
[18] |
AIoT
and edge-cloud for commercial vehicle traffic safety |
Enhanced safety and efficiency in commercial vehicle
operations |
AVs not primary focus |
Edge computing and federated learning for data
privacy |
Scalability and privacy in AIoT
systems |
|
[19] |
Strategic AI and IoT integration for smart city
transformation |
Enhanced urban operations and emergency response |
Limited AV focus, more on infrastructure
optimization |
Autonomous data analysis from IoT sensors |
Infrastructure costs and standardization gaps |
|
[20] |
IoT embedded in AVs with AI for intelligent mobility |
AVs improve traffic and environment al outcomes |
Direct focus on AV performance and sustainability |
Data collection and analysis via IoT sensors in AVs |
Infrastructure and communication protocol issues |
|
[21] |
AI and IoT in public transit system enhancement |
Predictive maintenance and route optimization |
AVs not central, focus on transit operations |
Real-time monitoring and dynamic scheduling |
Data privacy and operational complexity |
|
[22] |
Big data and AI algorithms for intelligent transport |
Optimized traffic planning and congestion management |
AVs included in ITS applications |
Large-scale data analytics and AI integration |
Data privacy and algorithm scalability |
|
[23] |
AIoT
innovations for sustainable transportation |
Reduced emissions and improved public transit |
AVs part of broader smart mobility solutions |
Environmental monitoring via IoT sensors |
Security and operational reliability concerns |
|
[24] |
ML and DL integrated with IoT for congestion
management |
Notable travel time and congestion reductions |
Autonomous vehicle integration in traffic systems |
Real-time sensor data and adaptive signal control |
Challenges in data integration and model training |
|
[25] |
Optimized network architecture for smart traffic
management |
Reduced vehicle delay and queue length |
AVs included in adaptive signal control |
Reinforcement learning with SUMO traffic data |
Network bandwidth and communication demands |
|
[26] |
Fusion of AI, IoT, V2X for intelligent transport
channels |
Enhanced traffic flow and reduced congestion |
AVs integrated for eco-friendly mobility |
Real-time data collection and analysis via IoT |
Privacy, cybersecurity, and infrastructure
challenges |
|
[27] |
AI-driven urban computing with IoT data integration |
Enhanced traffic management and urban planning |
Limited AV discussion, focus on urban computing |
Real-time data analytics for adaptive urban systems |
Data volume and computational challenges |
|
[28] |
IoT-enabled AI for traffic, safety, and parking
management |
Improved safety and traffic control |
AI supports driver monitoring and accident
prevention |
Sensor data fusion and AI classification |
Privacy and sensor deployment issues |
|
[29] |
AI and ML applications in intelligent transportation |
Traffic congestion and accident prevention |
Autonomous vehicles as part of safety systems |
Sensor data utilization for predictive models |
Data privacy and computational complexity |
|
[30] |
ML and IoT adaptive traffic management system |
Reduced congestion and travel time |
AVs integrated in mixed traffic scenarios |
Real-time data from vehicles and infrastructure |
Wireless communication and data processing limits |
|
[31] |
Big data, AI, and IoT in smart city applications |
Enhanced urban resource management and livability |
AVs discussed within broader smart city context |
Integrated data analytics for urban systems |
Policy and skills gap challenges |
|
[32] |
AI and big data for autonomous vehicle management |
Reduced waiting and travel times for AV services |
Direct focus on AV fleet optimization |
Network calculus and AI for queue modelling |
Complexity in mobility and service optimization |
|
[33] |
Big data and IoT for smart city traffic management |
Dynamic traffic signal adjustment and congestion
reduction |
AVs not primary focus |
Real-time sensor data and adaptive learning |
Data integration and system scalability |
|
[34] |
IoT and Deep Q Networks for traffic density
management |
Improved traffic flow and congestion control |
AVs indirectly supported via traffic optimization |
Reinforcement learning with IoT data fusion |
Scalability and real-time adaptation challenges |
|
[35] |
ML and IoT for smart transportation applications |
Route optimization and accident prevention |
AVs not
central |
ML algorithms for parking and traffic management |
Research gaps in parking and lighting systems |
|
[36] |
Edge-enabled IoT smart traffic management system |
Reduced congestion and emissions, improved mobility |
AVs indirectly supported via traffic optimization |
Edge computing for real-time data processing |
Latency and infrastructure deployment issues |
|
[37] |
AI framework integrating IoT traffic data for smart
cities |
Optimized traffic flow and congestion reduction |
AVs not primary focus |
Machine learning and deep learning for forecasting |
Data quality and model generalization |
|
[38] |
IoT-based smart traffic management with real-time
data |
Optimized traffic light control and emergency
response |
AVs not
central |
Real-time sensor and map data integration |
Data accuracy and system responsiveness |
3.2. Critical Analysis and Synthesis
The
reviewed literature collectively underscores the transformative potential of AI
and IoT technologies in urban mobility, particularly in traffic management,
public transportation, and autonomous vehicles. Its strengths include the
integration of real-time data analytics, predictive modeling,
and adaptive control systems that enhance efficiency and sustainability.
However, several studies have revealed methodological limitations, such as
reliance on historical data, scalability challenges, and insufficient attention
to privacy and ethical concerns. Furthermore, although technological
advancements are well documented, practical implementation and user acceptance
remain underexplored. The synthesis highlights the need for more robust,
interdisciplinary approaches that address infrastructural, regulatory, and
societal dimensions to fully realize the benefits of AI-IoT in urban mobility
(Tab. 2).
Tab. 2
The
transformative potential of AI and IoT technologies in urban mobility
|
Aspect |
Strengths |
Weaknesses |
|
Integration of AI and IoT in Traffic Management |
Many studies demonstrate effective use of AI-driven
predictive models combined with IoT sensor networks to optimize traffic flow
and reduce congestion, employing real-time data for dynamic signal control
and anomaly detection, which significantly improves urban mobility efficiency
and environmental outcomes [7]. The use of machine learning algorithms such
as reinforcement learning and deep learning enhances adaptability to changing
traffic conditions [34]. |
Despite promising results, several works highlight
limitations including dependency on historical data that may not capture
dynamic urban changes, leading to potential inaccuracies in predictions [11].
Scalability and computational complexity issues arise when deploying these
systems city-wide, and many models lack validation in diverse real-world
contexts [22]. Privacy concerns related to extensive data collection remain
insufficiently addressed. |
|
AI and IoT in Public Transportation Systems |
Research shows that AI and IoT integration enables
predictive maintenance, dynamic scheduling, and real-time passenger
information systems, which collectively improve reliability, reduce downtime,
and enhance user satisfaction. These technologies facilitate route
optimization and fare management, contributing to operational efficiency and
sustainability [23]. |
However, the literature reveals a scarcity of
large-scale empirical studies validating these systems in operational public
transit networks. Challenges such as infrastructure costs, interoperability
among heterogeneous devices, and user acceptance are often underexplored
[21]. Additionally, the complexity of integrating legacy systems with new
AI-IoT frameworks poses practical barriers. |
|
Autonomous Vehicles within AI-IoT Frameworks |
The convergence of AI, IoT, and autonomous vehicle
technologies is well documented to improve safety, traffic congestion, and
environmental impact through advanced sensors, V2X communication, and
data-driven decision-making [20]. Studies emphasize the potential of AVs to
enhance urban mobility and accessibility, especially for vulnerable
populations. |
Despite technological progress, significant
challenges persist regarding regulatory frameworks, cybersecurity, and public
trust [16]. Many studies focus on simulation or theoretical models with
limited real-world deployment data. Ethical considerations and human factors,
such as user behavior and acceptance, are
inadequately addressed. The integration of AVs into existing urban
infrastructure remains complex and costly. |
|
Data Analytics and Machine Learning Methodologies |
The application of advanced machine learning
techniques, including deep learning, reinforcement learning, and clustering
algorithms, has been shown to improve traffic prediction accuracy, incident
detection, and adaptive control. These methods enable proactive traffic
management and efficient resource allocation [34]. |
Methodological weaknesses include overfitting risks,
lack of explainability in AI models, and challenges in handling heterogeneous
and noisy data streams [12]. Many studies do not sufficiently address the
transparency and interpretability of AI decisions, which affects trust and
adoption [17]. Furthermore, the computational demands of complex models may
hinder real-time applicability in large-scale urban settings. |
|
Privacy, Security, and Ethical Considerations |
Some research acknowledges the critical importance
of data privacy, cybersecurity, and ethical issues in AI-IoT urban mobility
solutions, proposing encryption, multi-factor authentication, and blockchain
for data integrity [19]. The need for responsible AI adoption and public
engagement is emphasized. |
Nonetheless, these concerns are often treated
superficially or as secondary issues. There is a lack of comprehensive
frameworks addressing governance, data ownership, and ethical standards
across studies. The potential for surveillance and misuse of data remains a
significant barrier to widespread acceptance [13]. |
|
Infrastructure and Implementation Challenges |
Several papers highlight the necessity of robust
infrastructure, including edge computing, vehicular networks, and sensor
deployment, to support AI-IoT systems effectively [36]. Case studies
demonstrate improved traffic management and reduced emissions through such
implementations. |
However, infrastructural costs, interoperability
issues, and maintenance requirements pose significant challenges. Many
studies rely on simulations or pilot projects without extensive real-world
validation [15]. The integration of heterogeneous technologies and legacy
systems is complex, and there is limited discussion on scalability and
long-term sustainability [25]. |
|
User Acceptance and Societal Impact |
Research indicates growing acceptance of AI-driven
urban mobility solutions, with benefits including reduced travel time, fuel
consumption, and enhanced safety [10]. Studies also explore behavioral aspects and public perceptions, which are
crucial for adoption. |
Despite this, user acceptance remains a critical
challenge, with limited empirical data on societal impacts, equity, and
accessibility [21]. Resistance due to privacy concerns, trust deficits, and
lack of awareness is insufficiently addressed. The social implications of
automation and AI-driven decision-making require deeper investigation. |
4. DISCUSSION
The reviewed literature reveals several dominant
themes regarding the integration of AI and IoT in urban mobility, particularly
focusing on traffic management, public transportation systems, and autonomous
vehicles. The major themes include intelligent traffic management systems
leveraging real-time data and machine learning for congestion reduction and
safety improvements, IoT-enabled public transit optimization with predictive
maintenance and scheduling, and the deployment of autonomous vehicles within smart
city frameworks. Emerging areas examine challenges such as data privacy,
infrastructure complexities, and public acceptance, while highlighting the
convergence of AI, IoT, and edge computing to foster smarter, safer, and more
sustainable urban mobility ecosystems.
The evolution of research on AI and IoT in urban
mobility illustrates a clear progression from foundational concepts and early
technological integration to sophisticated, multi-dimensional smart city
applications. Initial studies focused on conceptual frameworks and basic
applications in traffic and public transit management, gradually advancing
towards predictive analytics, autonomous vehicle integration, and the fusion of
big data with AI and IoT. Recent studies have emphasized real-time adaptive
systems, edge computing, and the intricacies of data privacy and cybersecurity
in urban mobility solutions. This chronological overview highlights the growing
complexity and interdisciplinary nature of research aimed at creating
efficient, sustainable, and user-centered urban
transportation ecosystems.
The reviewed literature generally agrees on the
critical role of integrating AI and IoT technologies to enhance urban mobility,
particularly in traffic management, public transportation, and autonomous
vehicle deployment. Most studies highlight the improvements in congestion
reduction, travel time efficiency, and environmental sustainability achieved by
these technologies. However, there is divergence concerning the extent of the
impact of autonomous vehicles, the maturity of deployment strategies, and the challenges
related to privacy, infrastructure, and user acceptance. These differences
often arise from variations in the study focus, geographical context,
technological maturity, and methodological approaches.
5. THEORETICAL &
PRACTICAL IMPLICATIONS
5.1. Theoretical Implications
The
integration of AI and IoT in urban mobility advances existing theories of
intelligent transportation by demonstrating enhanced capabilities in real-time
data collection, predictive analytics, and adaptive traffic management. These
technologies support dynamic and responsive urban traffic systems that
challenge traditional static traffic control models [1]. The convergence of AI,
IoT, and machine learning fosters a multilayered understanding of urban
mobility, emphasizing the importance of interconnected systems and data-driven
decision-making. This supports the theoretical framework of smart cities as
complex adaptive systems that require the holistic integration of technologies
[31].
The
findings highlight the evolving role of autonomous vehicles within AI-IoT
ecosystems, extending theories on mobility by incorporating
vehicle-to-everything (V2X) communication and collaborative decision-making,
which enhance traffic flow and safety beyond individual vehicle autonomy [32].
The literature underscores challenges related to data privacy, cybersecurity,
and trust in AI outputs, which refine theoretical models by incorporating
socio-technical dimensions such as ethical considerations and human-machine
interaction in urban mobility systems.
Reinforcement
learning and deep learning applications in traffic management introduce novel
theoretical perspectives on adaptive control and optimization in complex urban
environments, suggesting a shift from rule-based to learning-based traffic
systems [25]. The concept of parallel intelligent transportation systems, which
combine real and artificial counterparts for simulation and prediction, expands
the theoretical understanding of transportation system evolution and planning
under uncertainty [3].
5.2. Practical Implications
AI-
and IoT-enabled traffic management systems offer practical solutions for
reducing congestion, optimizing traffic signal timings, and improving emergency
response times, which can significantly enhance urban mobility efficiency and
reduce environmental impact [36]. The deployment of AI-driven predictive
maintenance and dynamic scheduling in public transportation improves
operational reliability and passenger satisfaction, supporting sustainable
transit systems and reducing downtime [21]. Addressing ethical, privacy, and
community engagement challenges is critical for the successful implementation
of AI-IoT urban mobility solutions, necessitating inclusive policymaking and
transparent governance frameworks to build public trust and acceptance in the
technology.
The
integration of autonomous vehicles within IoT frameworks facilitates safer and
more efficient urban transport by enabling real-time communication with
infrastructure and other vehicles, which can inform policy development
regarding infrastructure investment and regulatory standards [20]. Edge
computing combined with AI-IoT architectures enhances the data processing speed
and reduces the latency in traffic management, offering scalable and
energy-efficient solutions that are suitable for large metropolitan areas [14].
The adoption of machine learning models for traffic flow prediction and anomaly
detection supports the design of adaptive traffic control systems that respond
to dynamic urban conditions, providing actionable insights for city planners
and traffic authorities [35].
6. LIMITATIONS OF THE
LITERATURE
Several
studies exhibit a geographic bias, predominantly focusing on developed regions
such as Europe and North America, which limits the external validity of the
findings for global urban contexts, especially in the Global South. This
restricts the generalizability of the conclusions to diverse urban
environments. Many studies depend heavily on simulations and theoretical models
rather than real-world deployments, which constrains the ecological validity of
the results. This methodological constraint may overlook the practical
challenges and contextual factors affecting implementation [30].
A
recurring limitation is the insufficient addressing of data privacy,
cybersecurity, and ethical issues related to the collection and use of large
amounts of urban mobility data. This gap undermines trust and may hinder the
widespread adoption of AI-IoT solutions [13]. The literature lacks extensive
longitudinal research to assess the sustained impact and evolving effectiveness
of AI and IoT interventions in urban mobility. This limits our understanding of
the long-term benefits and potential unintended consequences of this treatment
approach. Few studies have thoroughly investigated user attitudes, perceptions,
and behavioral responses to AI and IoT technologies
in urban mobility, which are critical for adoption and successful integration.
This gap affects the practical applicability of the proposed solution.
Many
studies acknowledge but do not empirically address the infrastructural and
scalability challenges inherent in deploying AI and IoT systems on a citywide
scale, limiting insights into feasibility and operational constraints [26]. The
literature often focuses on isolated aspects such as traffic management, public
transit, or autonomous vehicles without fully integrating these domains, which
limits a comprehensive understanding of urban mobility ecosystems [28]. Ethical
considerations, including fairness, bias, and the societal impacts of AI-IoT
integration in urban mobility, are underexplored, weakening the holistic
assessment of the consequences of these technologies on urban mobility.
Some
studies highlight challenges related to the computational demands and
scalability of AI and big data algorithms, which may impede real-time
application and widespread deployment in complex urban environments [22].
7. RESEARCH GAP &
FUTURE RESEARCH DIRECTION
The article presents the future research scope as
well as research gaps from the reviewed papers, as mentioned in the Table 3.
Tab. 3
The future research scope
|
Gap Area |
Description |
Future Research Directions |
Justification |
|
Dynamic Validation of AI-IoT Models in Real-World
Urban Contexts |
Many AI-IoT traffic management models rely heavily
on historical data and simulations, lacking validation in diverse, dynamic
urban environments. |
Conduct longitudinal field studies deploying AI-IoT
systems in multiple cities with varying traffic patterns to evaluate
real-time adaptability and robustness. Develop frameworks
for continuous model updating
with live data. |
Reliance on static or historical data limits model
accuracy and responsiveness to real-world changes, reducing practical
effectiveness [22]. |
|
Integration Challenges of Legacy Public Transit
Systems with AI-IoT |
Existing public transportation infrastructures often
lack compatibility with advanced AI-IoT frameworks, hindering seamless
integration. |
Develop middleware and standardized protocols to
enable interoperability between legacy transit systems and AI-IoT platforms. Pilot integration
projects focusing on scalability and cost-effectiveness. |
Infrastructure heterogeneity and legacy system
constraints impede adoption and limit operational improvements [21]. |
|
Regulatory and Ethical Frameworks for Autonomous
Vehicle Deployment |
Current regulatory policies and ethical guidelines
for AVs are insufficiently developed, creating barriers to deployment and
public trust. |
Formulate comprehensive, multi-stakeholder
regulatory frameworks addressing safety, liability, data privacy, and ethical
AI use in AVs. Include
public consultation and transparency
mechanisms. |
Regulatory uncertainty and ethical concerns
undermine AV adoption and integration into urban mobility [16]. |
|
Explain ability and Trust in AI Decision-Making for
Urban Mobility |
AI models used in traffic and AV management often
lack transparency, limiting user trust and acceptance. |
Research interpretable AI techniques tailored for
urban mobility applications. Develop user-centric interfaces that explain AI
decisions in accessible terms to stakeholders and the public. |
Trust and explainability are critical for societal
acceptance and responsible AI deployment [17]. |
|
Scalability and Computational Efficiency of AI-IoT
Systems |
Many AI-IoT solutions face challenges scaling to
city-wide deployments due to computational complexity and data volume. |
Investigate edge computing, distributed AI, and
federated learning approaches to reduce latency and computational load. Benchmark scalability
in large urban testbeds. |
Scalability constraints limit real-time
responsiveness and broad applicability of AI-IoT systems [36]. |
|
Privacy and Cybersecurity in AI-IoT Urban Mobility
Systems |
Data privacy and cybersecurity risks are
under-addressed, threatening user data and system integrity. |
Develop robust encryption, blockchain-based data
integrity solutions, and privacy-preserving AI algorithms. Establish governance
models for data ownership
and access control. |
Privacy breaches and cyberattacks can erode public
confidence and disrupt critical urban mobility services [19]. |
|
User Acceptance and Societal Impact of AI-IoT
Mobility Solutions |
Limited empirical research exists on societal
perceptions, equity, and behavioral impacts of
AI-IoT urban mobility technologies. |
Conduct mixed-methods studies on user attitudes,
accessibility, and equity implications. Design inclusive engagement
strategies to incorporate diverse community inputs in system design. |
Understanding societal impact is essential to ensure
equitable, ethical, and widely accepted mobility solutions [21]. |
|
Real-Time Multi-Source Data Fusion for Predictive
Urban Traffic Management |
Current models often struggle with heterogeneous
data integration from diverse IoT sensors and sources for accurate
prediction. |
Develop advanced multi-modal data fusion algorithms
combining IoT sensor data, social media, and vehicle telemetry. Validate predictive
accuracy in live urban environments. |
Effective data fusion enhances prediction accuracy
and system responsiveness but remains technically challenging [30]. |
|
Economic and Environmental Impact Assessment of
AI-IoT Urban Mobility |
There is a lack of comprehensive studies quantifying
long-term economic benefits and environmental trade-offs of AI-IoT
implementations. |
Perform longitudinal cost-benefit and life-cycle
environmental impact analyses of AI-IoT urban mobility projects. Include comparative
studies across different urban contexts. |
Quantitative impact assessments guide sustainable
investment and policy decisions [23]. |
|
Addressing Infrastructure Costs and Standardization
Gaps |
High infrastructure costs and lack of standardized
protocols hinder widespread AI-IoT deployment in urban mobility. |
Research cost-effective sensor deployment strategies
and develop international standards for AI-IoT interoperability in
transportation systems. Promote public-private partnerships for infrastructure
investment. |
Infrastructure affordability and standardization are
prerequisites for scalable, interoperable urban mobility solutions [25]. |
In the Table 4
is shown the strength and weakness of modern technologies.
Tab. 4
Strength and
weakness of modern technologies
|
Aspect |
Strengths |
Weaknesses |
|
Integration of AI and IoT in Traffic Management |
Many studies demonstrate effective use of AI-driven
predictive models combined with IoT sensor networks to optimize traffic flow
and reduce congestion, employing real-time data for dynamic signal control
and anomaly detection, which significantly improves urban mobility efficiency
and environmental outcomes [7]. The use of machine learning algorithms such
as reinforcement learning and deep learning enhances adaptability to changing
traffic conditions [24]. |
Despite promising results, several works highlight
limitations including dependency on historical data that may not capture
dynamic urban changes, leading to potential inaccuracies in predictions [11].
Scalability and computational complexity issues arise when deploying these
systems city-wide, and many models lack validation in diverse real-world
contexts [22]. Privacy concerns related to extensive data collection remain
insufficiently addressed. |
|
AI and IoT in Public Transportation Systems |
Research shows that AI and IoT integration enables
predictive maintenance, dynamic scheduling, and real-time passenger
information systems, which collectively improve reliability, reduce downtime,
and enhance user satisfaction. These technologies facilitate route
optimization and fare management, contributing to operational efficiency and
sustainability [21]. |
However, the literature reveals a scarcity of
large-scale empirical studies validating these systems in operational public
transit networks. Challenges such as infrastructure costs, interoperability
among heterogeneous devices, and user acceptance are often underexplored.
Additionally, the complexity of integrating legacy systems with new AI-IoT
frameworks poses practical barriers [19]. |
|
Autonomous Vehicles within AI-IoT Frameworks |
The convergence of AI, IoT, and autonomous vehicle
technologies is well documented to improve safety, traffic congestion, and
environmental impact through advanced sensors, V2X communication, and
data-driven decision-making [20]. Studies emphasize the potential of AVs to
enhance urban mobility and accessibility, especially for vulnerable
populations [10]. |
Despite technological progress, significant
challenges persist regarding regulatory frameworks, cybersecurity, and public
trust. Many studies focus on simulation or theoretical models with limited
real-world deployment data. Ethical considerations and human factors, such as
user behavior and acceptance, are inadequately
addressed [11]. The integration of AVs into existing urban infrastructure
remains complex and costly. |
|
Data Analytics and Machine Learning Methodologies |
The application of advanced machine learning
techniques, including deep learning, reinforcement learning, and clustering
algorithms, has been shown to improve traffic prediction accuracy, incident
detection, and adaptive control [35]. These methods enable proactive traffic
management and efficient resource allocation. |
Methodological weaknesses include overfitting risks,
lack of explainability in AI models, and challenges in handling heterogeneous
and noisy data streams [12]. Many studies do not sufficiently address the
transparency and interpretability of AI decisions, which affects trust and
adoption [17]. Furthermore, the computational demands of complex models may
hinder real-time applicability in large-scale urban settings. |
|
Privacy, Security, and Ethical Considerations |
Some research acknowledges the critical importance
of data privacy, cybersecurity, and ethical issues in AI-IoT urban mobility
solutions, proposing encryption, multi-factor authentication, and blockchain
for data integrity. The need for responsible AI adoption and public
engagement is emphasized [16]. |
Nonetheless, these concerns are often treated
superficially or as secondary issues. There is a lack of comprehensive
frameworks addressing governance, data ownership, and ethical standards
across studies. The potential for surveillance and misuse of data remains a
significant barrier to widespread acceptance [8]. |
8. CONCLUSION
This study demonstrates that the proposed
integration of artificial intelligence and Internet of Things frameworks can
significantly improve urban mobility through optimized traffic management,
intelligent public transportation systems, and enhanced readiness for the
deployment of autonomous vehicles. Here, we have reviewed and identified
research gaps and future directions from the cited articles. These findings
confirm that AI-driven predictive analytics and adaptive control models
effectively reduce congestion, minimize delays, and contribute to lower carbon
emissions compared with conventional traffic management approaches.
With respect to autonomous vehicles, our analysis
highlights the potential and limitations of IoT-enabled decision-making. While
this is consistent with global reports on improved safety and congestion
reduction, we emphasize that infrastructural readiness and regulatory clarity
remain the most pressing barriers to the real-world implementation of these
vehicles. In contrast to simulation-based investigations common in the
literature, our study incorporates empirical deployment-level constraints,
providing a more realistic perspective on the challenges of adoption.
Additionally, our findings show that edge computing
and big data integration enhance the real-time responsiveness of the system.
However, our work underscores unresolved issues in data heterogeneity, privacy,
and cybersecurity, where prior studies often provide only conceptual
discussions without governance-oriented frameworks. Finally, while the
literature broadly recognizes the societal dimension, our research stresses the
role of public trust, explainability, and user-centric system design as non-trivial
determinants of sustainable adoption. Several studies have emphasized the
fusion of AI, IoT, and V2X or vehicle-to-infrastructure communication to
enhance intelligent transportation systems and autonomous vehicle support [26].
In conclusion, this study not only supports but
also extends the global evidence on AI-IoT–enabled mobility. By addressing
scalability, interoperability, and socio-ethical challenges, this study
contributes to bridging the gap between technological innovation and practical
urban deployment.
References
1.
Hossain M., H. Khalid, A.P. Rao, M. Lootah, S.S.K. Al-Mohammedi, S.R. Majeed. 2024. “Comprehensive
review of AI, IoT, and ML in enhancing urban mobility and reducing carbon
footprints.” IEEE, Third International Conference on Sustainable Mobility
Applications, Renewables and Technology (SMART). Dubai, United
Arab Emirates, DOI: 10.1109/SMART63170.2024.10815521.
2.
Dia H. 2016. “The real-time city: Unlocking the
potential of smart mobility.” Australasian
Transport Research Forum 2016 Proceedings, 16-18 November 2016,
Melbourne, Australia.
3.
Zhu F., Y. Lv, Y. Chen, X.
Wang, G. Xiong, F.Y. Wang. 2020. “Parallel transportation systems: Toward
IoT-enabled smart urban traffic control and management.” IEEE Transactions on Intelligent Transportation
Systems 21(10): 4063-4071. DOI:
10.1109/TITS.2019.2934991.
4.
Yuan T., W. Neto, C. Rothenberg, K. Obraczka, C. Barakat, T. Turletti.
2022. “Machine learning for next-generation intelligent transportation systems:
A survey.” Transactions on Emerging Telecommunications Technologies 33(4): 1-35. DOI:10.1002/ett.4427.
5.
Elassy M., M. Al-Hattab, M.
Takruri, S. Badawi. 2024. “Intelligent transportation systems for sustainable
smart cities.” Transportation Engineering 16: 1-18.
DOI: 10.1016/j.treng.2024.100216.
6.
Dündar M. 2024. “The integration of IoT, AI, and
machine learning in urban systems.” Next Frontier for Life Sciences and AI 8(1): 185-190. DOI: 10.62802/860ded41.
7.
Abbasova S. 2024. “Navigating the gridlock: Innovative
strategies for traffic management and control.” Luminis Applied
Science and Engineering 1(1): 84-91. DOI:
10.69760/lumin.202400007.
8.
Kushwaha A. 2024. “Application of AI and IoT in
traffic management of large metropolitan cities.” International Journal of
Innovative Science and Research Technology 9(4): 3426-3431. DOI: 10.38124/ijisrt/IJISRT24APR2069.
9.
Pali P., S. Verma, A. Patel, V. Soni. 2023.
“Intelligent urban transportation systems: A survey of AI-driven
innovations and future directions.” International Journal of Innovative
Research in Science, Engineering and Technology 12(5): 8068-8073.
10.
Makahleh H.Y., E.J.S.
Ferranti, D. Dissanayake. 2024. “Assessing the role of autonomous vehicles in
urban areas: A systematic review of literature.” Future Transportation
4(2): 321-348. DOI: 10.3390/futuretransp4020019.
11.
Englund C., E.E. Aksoy, F. Alonso-Fernandez, M.D.
Cooney, S. Pashami, B. Åstrand.
2021. “AI perspectives in smart cities and communities to enable road vehicle
automation and smart traffic control.” Smart Cities 4(2): 783-802. DOI: 10.3390/smartcities4020044.
12.
Mirindi D. 2024. “A review of the advances in
artificial intelligence in transportation system development.” Journal of
Civil, Construction and Environmental Engineering 9(3): 72-83. DOI:
10.11648/j.jccee.20240903.13.
13.
Dovzhenko N., N. Mazur, Y.V. Kostiuk, S. Rzaieva. 2024. “Integration of IoT and artificial
intelligence into intelligent transportation systems.” Cybersecurity Education Science Technique 2(26): 430-444.
DOI: 10.28925/2663-4023.2024.26.708.
14.
Chavhan S., D. Gupta, S.P. Gochhayat,
B.N. Chandana, A. Khanna,
K. Shankar, J.J.P.C. Rodrigues.
2022. “Edge computing
AI-IoT integrated energy-efficient intelligent transportation system for smart
cities.” ACM Transactions on
Internet Technology. 22(4): 1-18. DOI: 10.1145/3491210.
15.
Limkar S., W.V. Ashok, P. Shende, K. Wagh, S.K. Wagh, A. Kumar. 2023. “Intelligent transportation system using
vehicular networks in the Internet of Vehicles for smart cities.” Journal of Electrical Systems 19(2): 58-67. DOI: 10.52783/jes.691.
16.
Janeera D.A., S.S.R. Gnanamalar, K.C. Ramya, A.G.A. Kumar. 2021. “Internet of Things and
artificial intelligence-enabled secure autonomous vehicles for smart cities.” Automotive
Embedded Systems. Innovations in Communication and Computing. Springer.
DOI: 10.1007/978-3-030-59897-6_11
17.
Kumar A., N. Batra, A. Mudgal, A.L. Yadav. 2024.
“Navigating urban mobility: A review of AI-driven traffic flow management in
smart cities.” Proceedings of the 11th International Conference on
Reliability, Infocom Technologies and Optimization (ICRITO). Noida, India.
18.
Subbiah A., A. Mahfoud, A. Kumari. 2024. “Smart urban
traffic management: Leveraging automatic control and intelligent systems for
improved safety in commercial vehicle road banning operations.” Proceedings
of the IEEE 6th Symposium on Computers & Informatics (ISCI). Kuala Lumpur, Malaysia. DOI:
10.1109/ISCI62787.2024.10668243
19.
Hoang T.V. 2024. “Impact of integrated artificial
intelligence and Internet of Things technologies on smart city transformation.”
Journal of Technical Education Science. 19(1): 64-73. DOI: 10.54644/jte.2024.1532.
20.
Singh B., C. Kaunert, M.K. Arora, S. Lal. 2025.
“Autonomous transportation and smart vehicles enhancing mobility solutions: Fueling IoT solutions achieving SDG 11.” Sustainable
Cities and Future of Urban Development. IGI Global. New York, USA. DOI: 10.4018/979-8-3693-6740-7.ch003.
21.
Vemuri N., V.M. Tatikonda,
N. Thaneeru. 2024. “Enhancing public transit system
through AI and IoT.” International Journal of Scientific Research and
Management 12(2): 1057-1071. DOI: 10.18535/ijsrm/v12i2.cs03.
22.
Abirami S., M. Pethuraj, M. Uthayakumar, P. Chitra.
2024. “A systematic survey on big data and artificial intelligence algorithms
for intelligent transportation system.” Case Studies on Transport Policy 17: 100-110. DOI: 10.1016/j.cstp.2024.04.012.
23.
Alaba F.A., A. Oluwadare, U. Sani,
A.A. Oriyomi, A.O. Lucy, O. Najeem.
2024. “Enabling
sustainable transportation through IoT and AIoT
innovations.” Intelligent Technologies and Robotics. Springer, Cham, Switzerland.
DOI: 10.1007/978-3-031-56854-7_18.
24.
Khan A., P. Ivan. 2023. “Integrating machine learning
and deep learning in smart cities for enhanced traffic congestion management:
An empirical review.” Journal of Urban Development and Management 2(4): 211-221.DOI: 10.56578/judm020404.
25.
Sasikumar D., G. Sriram, K. Nelson, P. Harish. 2023.
“Smart traffic management system through optimized network architecture for the
smart city paradigm shift.” International Conference on Intelligent Systems
for Communication, IoT and Security (ICISCoIS). Coimbatore, India. DOI:
10.1109/ICISCoIS56541.2023.10100338.
26.
Ullah U., M. Usama, Z. Muhammad, A. Akbar, S. Latif,
R. Ullah. 2024. Artificial Intelligence for Intelligent Systems. Fundamentals,
Challenges, and Applications. Taylor & Francis. DOI:
10.1201/9781003496410.
27.
Haldorai A., B.R. Lincy, S.
Murugan, M. Balakrishnan. 2024. Artificial Intelligence for Sustainable
Development. Springer, Cham, Switzerland. DOI: 10.1007/978-3-031-44841-2_12.
28.
Nithya K., S. Mythili, M. Kalamani, M. Krishnamoorthi. 2024. “AI approaches
in intelligent transportation systems.” In: Artificial Intelligence for
Future Intelligent Transportation. Smarter and Greener Infrastructure
Design. Taylor & Francis. DOI: 0.1201/9781003408468.
29.
Gangwani D., P. Gangwani. 2021. “Applications of
machine learning and artificial intelligence in intelligent transportation
system: A review.” Lecture Notes in Electrical
Engineering. Springer. DOI: 10.1007/978-981-16-3082-6_19.
30.
Lilhore U.K., A.L. Imoize, C.T. Li, S.
Simaiya, S.K. Pani, N. Goyal,
A. Kumar, C.C. Lee. 2022. “Design and implementation of an ML and IoT-based
adaptive traffic-management system for smart cities.” Sensors 22(8): 1-26. DOI: 10.3390/s22083100.
31.
Yao Y. 2022. “A review of the comprehensive
application of big data, artificial intelligence, and Internet of Things
technologies in smart cities.” Journal of Computational Methods in
Engineering Applications 2(1): 1-10. DOI: 10.62836/jcmea.v2i1.0004.
32.
Cui Q., Y. Wang, K.C. Chen, W. Ni, I.C. Lin, X. Tao.
2019. “Big data analytics and network calculus enabling intelligent management
of autonomous vehicles in a smart city.” IEEE
Internet of Things Journal. 6(2): 2021-2034. DOI: 10.1109/JIOT.2018.2873157.
33.
Nagalapuram J., S.
Samundeeswari. 2024. “A framework for smart city traffic management utilizing
BDA and IoT.” Engineering, Technology & Applied Science
Research 14(6): 18989-18993.
DOI: 10.48084/etasr.8003.
34.
Ranganathan C.S., L.N. Jayanthi, J.J. Babu, M. Manikandan, N. Sivakamy. 2024. “Urban mobility optimization with IoT and deep Q
networks for traffic density management.” 3rd International Conference for
Innovation in Technology (INOCON). Bangalore, India. DOI: 10.1109/INOCON60754.2024.10511802.
35.
Mukhopadhyay S., A. Kumar, J. Gupta, A. Bhatnagar,
M.P. Kantipudi, M. Singh. 2024. “A review and analysis of
IoT-enabled smart transportation using machine learning techniques.” International
Journal of Transport Development and Integration 8(1): 61-77. DOI: 10.2495/TDI-V8-N1-61-77.
36.
Jabakumar A.K. 2023.
“Edge-enabled smart traffic management system: An IoT implementation for urban
mobility.” Research Journal of Computer
Systems and Engineering 4(2): 160-173. DOI:
10.52710/rjcse.85.
37.
Moumen I., J. Abouchabaka,
N. Rafalia. 2023. “Enhancing urban mobility:
Integration of IoT road traffic data and artificial intelligence in smart city
environment.” Indonesian Journal of Electrical Engineering and Computer
Science 32(2): 1010-1017. DOI: 10.11591/ijeecs.v32.i2.
38.
Taiwo A.A., C.C. Nzeanorue,
S.O. Ayanwunmi, O.O. Ajiboye, A.I. Azeez, S. Hakeem,
C.G. Nzeanorue, J..C. Agba, P.D. Fakoyede, E. Enabulele,
V.I. Stephen, A. Oyesanya, M. Ogbe, R.A. Olusola.
2024. “Intelligent transportation system leveraging Internet of Things (IoT)
technology for optimized traffic flow and smart urban mobility management.” World Journal of Advanced Research and Reviews 22(3): 1509-1517. DOI: 10.30574/wjarr.2024.22.3.1886.
Received 17.07.2025; accepted in revised form 28.10.2025
Scientific Journal of Silesian
University of Technology. Series Transport is licensed under a Creative
Commons Attribution 4.0 International License