ABSTRACT Unregulated parking search is a major source of urban inefficiency, leading to unnecessary cruising, wasted time, driver frustration and increased emissions. Moreover, when drivers select spots solely based on immediate self‐interest, ignoring others' dynamic needs and preferences, this behaviour exacerbates congestion, conflict and overall discomfort. To address these challenges and advance towards a smart, sustainable and user‐centric mobility system, this paper proposes an Optimal Parking Allocation Scheme (OPAS). By intelligently integrating user behaviour and real‐time demand, OPAS dynamically assigns parking spaces that enhance overall system efficiency while flexibly providing each user with an optimal spot, balancing collective sustainability with individual comfort. The proposed OPAS wirelessly receives the user information upon arrival at the entrance and determines the best‐suited spot by solving an adaptive optimisation problem considering the travel and parking behaviour in terms of the number of passengers in the car, their walking and driving distances, the priority level of users, in addition to trends of the occupation ratio. The proposed OPAS dynamically tunes its objective parameters to align with overarching system goals, yielding successive, situation‐adaptive decisions. The performance of the proposed scheme is evaluated using data from an indoor parking model of a shopping centre in Gunma City, Japan. The results demonstrate that OPAS achieves a significant reduction in both individual walking distance and vehicular travel distance within the parking facility. Consequently, this efficiency directly translates to a decrease in total fuel consumption and associated emissions. Specifically, the proposed strategy achieves a daily reduction of 23 kg in emissions and 10 L in fuel consumption, highlighting the practical significance of the proposed approach.
ABSTRACT Tourism places increasing pressure on coastal cities, where seasonal congestion, fragmented urban development and changing citizen expectations challenge the sustainability of services and governance. To address these issues, we present TorreviejaSensing, a social sensing framework that transforms user‐generated content into actionable evidence for policy and planning. The system combines multilingual natural language processing over X (Twitter) posts and Google Maps reviews, a policy‐aware mapping of social signals to municipal strategic axes and a retrieval‐augmented generation (RAG) chatbot that provides transparent evidence‐grounded answers to policy queries. Applied to Torrevieja (Spain), the framework enables real‐time tracking of mobility and cleanliness complaints, early detection of seasonal pain points and alignment of digital discourse with institutional strategies. A mixed‐methods evaluation—including manual annotation of NLP outputs, baseline comparison and a pilot human assessment of chatbot responses—demonstrates both robustness and practical utility. Findings reveal strong seasonal dynamics, recurrent topic clusters consistent with known governance challenges and improved transparency of the RAG assistant compared to keyword‐only approaches. Beyond the Torrevieja case, the framework is low‐cost, transferable and provides a replicable model for integrating citizen discourse into sustainable urban tourism governance.
This study proposes an integrated interdisciplinary framework that combines AI tools with drone photogrammetry to address technical, theoretical and pedagogical dimensions of urban analysis. It uses high-precision geospatial data (n = 737) to validate urban complexity theories and professional competencies. The methodology involves orthomosaic generation, point clouds and 3D models using photogrammetry techniques. AI core methods, such as machine learning and computer vision, specifically automated feature detection, Structure from Motion (SfM) and Multi-View Stereo (MVS) algorithms embedded in commercial photogrammetry pipelines, supported spatial analysis, improving data quality and analytical capacity. The results demonstrate that automated semantic segmentation facilitates classification of point clouds for urban modelling. Automated feature-matching and lighting normalisation achieved a mean reprojection error of 0.102 pixels, yielding a high-fidelity dataset capable of capturing informal urban morphologies that remain unresolved in standard, publicly available satellite imagery. Beyond technical outcomes, a motivation survey of participating students revealed strong intrinsic and extrinsic engagement, while fostering competencies to address complex urban challenges. The study contributes to a technical workflow for AI-enhanced spatial reconstruction, empirical validation of urban complexity and fractal theories and organisation, and an exploratory pedagogical study on emerging technologies. Overall, the framework improves spatial precision, supports evidence-based urban governance and contributes to the UN's SDG goal 11.
ABSTRACT Urban paving and resurfacing operations frequently bury utility access points, complicating infrastructure maintenance and emergency response. Although ground‐penetrating radar (GPR) can be used to locate buried manholes, manual radargram interpretation requires specialised expertise and is time‐consuming and subjective. This paper introduces a systematic comparative evaluation of three YOLO (You Only Look Once) variants of different architectural generations, including YOLOv5s (anchor‐based), YOLOv8s (anchor‐free with decoupled heads), and YOLOv11s (transformer‐enhanced). We trained and validated these models using 9‐fold cross‐validation on 54 controlled radargrams obtained at 1000 MHz under systematically varied conditions: three surface types (asphalt, concrete, sand), six burial depths (0.05–0.75 m), and three compaction levels (C0, C5, C10). The results show that YOLOv5s showed the best performance under sparse data conditions, with accuracy = 0.7865, precision = 0.8198, recall = 0.7533, mAP@0.5 = 0.7950 and the lowest cross‐fold variability (RMSE = 0.0725). YOLOv8s achieved higher mAP@0.5:0.95 (0.3711), indicating good generalisation on strict IoU thresholds, but YOLOv11s showed high instability (RMSE = 0.2200), indicating overfitting. Paired t‐tests showed no statistically significant difference between architectures at α = 0.05, but practical stability differences that are deployment‐relevant were prominent. Detection accuracy degraded systematically at high compaction (C10) and at greater burial depths, consistent with electromagnetic attenuation in dense media. YOLOv5s also offered the shortest inference time, which supports near‐real‐time deployment scenarios. A prototype of the interface based on the LINE is used to demonstrate the technical feasibility of field workflows. These results suggest that the lighter YOLO architecture provides better stability and efficiency for GPR‐based utility detection when training data is limited. This can guide the model selection for infrastructure inspection applications.
ABSTRACT This paper presents a formally verified, lightweight security protocol for energy‐constrained IoT devices operating in regulated environments, with a primary focus on wearable medical systems. The protocol is designed to meet emerging cybersecurity requirements from HIPAA, GDPR and Health Canada and combines Ed25519‐based mutual authentication, ephemeral Curve25519 key exchange and ChaCha20‐Poly1305 authenticated encryption to provide secure, in‐transit protection across heterogeneous nodes with minimal computational and energy overhead. Formal verification using ProVerif and Scyther establishes resilience against key compromise, replay and session confusion, with guarantees of end‐to‐end authentication, forward secrecy, nonce freshness and key integrity in multi‐node settings. Practical feasibility is evaluated through a dual‐phase methodology. A Python‐based simulation framework examines runtime behaviour under adversarial and fault‐prone conditions, enabling analysis of implementation‐level state transitions and recovery behaviour beyond symbolic models. The protocol is then deployed on an STM32L431‐based ECG wearable relayed through a mobile gateway to a cloud server, representing a CPU‐bound short‐range IoT platform. Energy profiling using Nordic PPK2 shows a daily protocol overhead of approximately 7.91 mWh, under 16% of the system energy budget, supporting over 30 days of secure operation per charge. Complementary evaluation on a cellular IoT platform with hardware‐backed security further demonstrates that regulation‐ready secure communication is feasible across heterogeneous low‐power IoT deployments without compromising runtime efficiency.
In smart city environments, public safety increasingly depends on intelligent surveillance systems that can be capable of adapting to dynamic and context‐dependent access restrictions. Traditional systems often rely on static and predefined boundaries that fail to respond to rapidly changing environments such as construction sites, public gatherings or emergency situations. This paper introduces a novel deep learning‐driven framework using ground‐plane homography for real‐time proactive intrusion prediction within these dynamically restricted zones (DRZs). Our method first employs deep learning to accurately detect and localise physical restriction markers (e.g., traffic cones). We then utilise ground‐plane homography estimation to accurately map these markers into two‐dimensional ground‐plane perspective, precisely defining the spatial boundaries of the DRZ in real‐time. After the reactive detection of restriction markers region, intrusion prediction is achieved through sophisticated human trajectory analysis and future path extrapolation. By forecasting a person's path and identifying projected future presence within the dynamic ground‐plane zone, the system assists proactive alerts and adaptive security responses before an actual violation. To the best of our knowledge, this is the first system capable of predicting intrusions into areas dynamically demarcated by visual restriction markers. The experimental results on real‐world surveillance datasets demonstrate the system's effectiveness in identifying the presence of humans in DRZ, validating its potential for deployment in smart cities and critical infrastructure.
Understanding associations between sustainable development and the smart city is essential to achieve sustainable smart cities. Dubai has emerged as an example in the Middle East for adopting smart technologies to enhance urban living, with initiatives ranging from digital governance to intelligent transportation systems. However, the associations between sustainable development and smart city implementation in Dubai is limited. This study aims to investigate the application of the smart city in Dubai, assessing the smart city implementation in terms of sustainable development by applying the system archetypes to assess the implementation of the smart city in Dubai in terms of sustainable development issues. After the identification of the system archetypes, it is found that the implementation of smart city initiatives such as e-government and clean transportation are in line with sustainable development issues such as low-carbon emissions and less air pollution. Moreover, this study shows that the structure of the Limits to Growth archetypes has dominated the smart city development in Dubai. This means that the development of Dubai has had critical issues such as persistent traffic congestion and a polluted atmosphere. The findings stress that the smart city application in Dubai is a good exemplar that the smart city can be a sustainable smart city altogether. The second point is although the smart city enables us to achieve a sustainable smart city, the implementation of the smart city should be monitored regularly, especially if reinforcing loops dominate balancing loops as seen in the case of traffic congestion. This study contributes to enhancing the decision-making process of policymakers, industry stakeholders, government authorities and business managers regarding the implementation of smart initiatives as well as for city planners to achieve a sustainable smart city in other regions.
In large-scale deployments, the Internet of things (IoT) and wireless sensor networks (WSNs) often face challenges in transmitting collected data to the base station due to limited network coverage. Unmanned aerial vehicles (UAVs) can extend this coverage by flying to remote WSN areas and communicating with aggregator nodes (CH-nodes) to retrieve data. Designing UAV-assisted data collection systems therefore requires a careful consideration of both UAV and WSN constraints. This article proposes an energy-efficient approach for UAV-based data collection in IoT/WSNs. The problem is formulated to jointly optimise system cost and energy consumption while accounting for communication power, mission duration, and data importance. The solution proceeds in two steps. First, aggregator nodes are selected using clustering based on residual energy and inter-node distances to minimise system costs. Second, the UAV trajectory is generated using a Lévy flight strategy that follows the positions of the selected aggregators. Although this trajectory may be slightly longer than that produced by a deterministic TSP route, it increases the amount of collected data and prolongs both UAV and WSN lifetime by ensuring timely visits to distant cluster heads. Simulation results confirm the efficiency and robustness of the proposed method compared with existing solutions.
Mapping socio-ecological dynamics reveals how human and natural systems interact over time, supporting informed planning and balanced regional development. By detecting patterns in these interactions, mapping supports policymakers in navigating complex transitions and guiding sustainable regional planning. These transitions are particularly evident in regions experiencing synchronised coal mine and coal-fired power plant closures. Despite ongoing rehabilitation efforts worldwide, few studies explore how socio-ecological factors interact and evolve or employ mapping as an integrative tool. Directly addressing this gap, this study innovatively introduces a framework that treats coal mines and power plants as a connected nexus, analysing their regional impacts through an integrated mapping approach. This framework combines geospatial mapping with exploratory, causal, and predictive modelling to analyse spatiotemporal shifts in post-mining landscapes. Applied to the Latrobe Valley in Australia, the framework reveals the closure caused sharp declines in income and nighttime light intensity, with no immediate recovery in native vegetation. Projections indicate that without early intervention, the Valley risks deepening regional socioeconomic decline. Translating multifaceted data into an analytical format enables stakeholders to see through complexity, understand interconnected socio-ecological dynamics across phases, and coordinate governance to manage regional changes for balanced development strategies.
Smart cities, characterised by their extensive use of IoT devices, aim to enhance urban living through improved efficiency, sustainability and quality of life. However, the widespread integration of IoT technology introduces significant cybersecurity challenges, including vulnerabilities to cyberattacks, data breaches and infrastructure resilience issues. Addressing these challenges is critical to realising the full potential of smart city initiatives. Intrusion detection systems (IDS) play a vital role in safeguarding smart city environments. Numerous studies have explored various IDS methodologies, yet the dynamic and complex nature of smart city IoT networks demands continuous advancements. In this article, we present a novel IDS approach that leverages machine learning techniques to enhance the detection and prevention of cyber threats in smart cities. Utilising the UNB CIC IoT 2023 Dataset, we develop and evaluate multiple models, including Random Forest Classifier, Decision Tree Classifier, KNN and AdaBoost. Our proposed IDS framework emphasises real-time threat detection ensuring both low latency and high accuracy. Through comprehensive data preprocessing and rigorous model training, our system demonstrates exceptional performance in identifying and neutralising cyber threats. The findings from this research reveal significant improvements in the security and privacy of smart city IoT infrastructures highlighting the effectiveness of integrating advanced AI methodologies.
This paper presents a hybrid human-centred sustainable smart city framework that integrates three key dimensions: Sustainability, Health & Wellbeing and Technologies. The proposed framework uses a hybrid index to evaluate smart city performance across these dimensions, applying a quantitative methodology based on min-max normalisation. By conducting an in-depth comparative analysis of smart city initiatives in Hong Kong and Riyadh, the study illustrates the adaptability of the hybrid framework in distinct urban contexts. Hong Kong's efforts in integrating (Internet of Things) IoT-based monitoring systems, digital health platforms, and active transportation highlight the focus on health and wellbeing in space-constrained urban environments. In contrast, Saudi Vision 2030 emphasises renewable energy deployment, large-scale smart infrastructure projects, and sustainability measures in Riyadh. Despite the differences in economic and regulatory environments, both cities demonstrate the effectiveness of embedding human-centred principles in urban development. The findings underscore that a balanced integration of technology, sustainability and wellbeing is essential for building resilient, inclusive, and sustainable smart cities, serving as a model for future urban development globally.
The smart city concept integrates various collaborative services to enhance urban living. However, these services introduce significant security concerns, especially in authentication, authorisation, and access control (AAA). To address these security challenges, researchers must design and implement frameworks that safeguard data exchange between smart services. This paper offers a taxonomy-based review of current solutions, focusing on how emerging technologies like Blockchain, artificial intelligence (AI), quantum computing, and hybrid approaches address AAA concerns. We evaluate these technologies based on key factors such as confidentiality, integrity, availability (CIA), trust, privacy, and scalability. Through a systematic review of literature from 2017 to 2024, we classify and assess methods that strengthen authentication, optimise access control, and refine authorisation processes to mitigate risks in data sharing. A major contribution of this paper is the integration of case studies, demonstrating real-world applications of these technologies in smart city contexts. Additionally, we explore the applicability of these solutions, highlighting their challenges and future potential. This research also outlines future directions for building secure, efficient, and scalable smart city ecosystems, ultimately facilitating the development of adaptable frameworks for smart city services.
Statistics indicate that many road accidents stem from driver negligence, such as collisions with parked cars, motorcyclists or pedestrians in blind spots. Although numerous studies address bus driver behaviour, most focus on freeways rather than urban streets. This paper introduces a safety assessment system (SAS) utilising existing onboard cameras on buses, eliminating the need for additional sensors. The SAS evaluates city bus drivers' behaviour when entering and exiting bus stops (ELBS), considering three key factors: entry velocity, head turn frequency to assess surroundings, and estimated distance from the road boundary upon arrival (DBR). Leveraging image and global positioning system (GPS) data, the system establishes risk levels for each scenario, aiding in identifying safe driving practices. To validate its feasibility, a professional survey was conducted, confirming alignment between the designed scoring scale and survey results.
The smart city framework has become a key approach to addressing urbanisation challenges over the last 2 decades. While KPIs have been developed for various smart city dimensions, security and safety remain underexplored. This paper addresses this gap through a systematic review of KPIs. The study examines how urban security and safety smartness is assessed, focusing on three questions: (RQ1) What indicators measure urban security and safety smartness? (RQ2) In which smart city dimensions are these KPIs located? (RQ3) How are these KPIs defined and quantified? Using PRISMA guidelines, databases including Web of Science, Scopus, and IEEE Xplore were searched, yielding 2369 sources. After screening, 38 studies were analysed. A total of 182 unique KPIs were identified and categorised into crime prevention and control (53), perceptions of safety (11), emergency and disaster management (50), and cybersecurity (68). Most KPIs focus on city outcomes, with fewer addressing smart technology functionalities. Definitions and measurement approaches lack consensus. This review identifies gaps in defining and measuring smart urban security and safety. Standardising KPIs and incorporating technology-specific metrics are key directions for future research.
The interest in smart city initiatives is continuously growing, and so is the interest in measuring their smartness and sustainability. The value of a Smart City is proved by comparing its results with sustainability goals and by measuring the benefits delivered to stakeholders, citizens and city authorities. For this reason, supporting the monitoring and evolution of smart cities is a challenging and crucial task. This can be accomplished by using key performance indicators (KPIs), which, in turn, inform the decision-making processes of smart city governance. However, ethical implications in KPIs evaluation may be hidden within the measurement process. For example, low trustworthiness of data sources used to collect KPIs input parameters can affect evaluation results without KPIs experts being aware of these implications. Ethical implications that could represent a perceived violation of an ethical value should be highlighted when KPIs results are proposed to stakeholders. In this paper, we have integrated an ethical risk traceability mechanism into MIKADO , a tool that we designed for assessing KPIs in smart cities. This mechanism involves both KPIs and ethics experts in the evaluation process, ensuring that the KPI assessment results—presented through dashboards obtained through code generation—include warnings about potential ethical issues. The extended framework has been evaluated by experts in the field during a dedicated focus group, who confirmed its usefulness in identifying ethical implications in the evaluation of smart city KPIs, with the aim of tracing ethical risks, and supporting ethics experts in carrying out this task.
Digital twinning is an advanced technology that involves creating virtual replicas of various physical systems. In smart cities, digital twins serve as digital representations that model and simulate various urban elements, such as environment protection (e.g., air quality), critical infrastructure, transportation networks and other urban management processes. It has recently gained considerable attention for its transformative potential, enabling city authorities to visualise and analyse complex city dynamics for better-informed decision-making. Therefore, this paper proposes a simplified layered architecture for smart city digital twins. The layers of the proposed architecture cover the range of operations required by the functionality of the digital twin and the interaction between them, from data transfer or synthesis to big data streaming and intelligent analytics. The paper also introduces an open-source software tool that realises the proposed architecture, with each layer designed as an independent Python module for easy integration and maintenance. Three case studies are used to demonstrate the capabilities of the tool. One use case addresses short-term forecasting of the air quality index, whereas the other use case targets the detection of an individual's respiratory condition based on data received from wearable devices. The third case combines the other two cases to offer a warning system for residents with medical conditions based on air quality. The results of the case studies show the tool's ability to effectively handle environment and eHealth-related use cases and combine them for the welfare of smart city residents, leading to a more resilient health-focused urban landscape.
Urban environments often pose challenges for individuals with mobility impairments due to inadequate pedestrian infrastructure. In addition, the lack of accurate mapping of accessibility features limits the ability to monitor and address these constraints effectively. This paper introduces a framework for Automating City Accessibility Mapping using AI (ACAMAI), that is, provides an AI-assisted pipeline for the automated identification and geolocation of urban accessibility constraints using Google Street View (GSV) panoramas. The ACAMAI pipeline comprises two main stages: (i) training a YOLOv8 object detector to recognise accessibility-related features, such as curb ramps, missing ramps, obstacles and surface problems, in 2D sidewalk images; and (ii) scanning 360 degrees GSV panoramas by extracting multiple perspective views to be analysed by the trained model. The model was trained on a combination of international (Project Sidewalk Dataset-PSD) and local (Porto Dataset-PTD) datasets, achieving high performance across classes, including 91% recall and 85% precision for curb ramps. In the panorama scanning stage, using a fine angular iterative step (2 degrees) maximised the recall, reaching 90% for curb ramps and 93% for obstacles in a locally annotated dataset (GSV Panorama Porto Dataset-GSV-PPD). Although this improved detection coverage, it also led to a high number of redundant predictions, which contributed to a reduced overall precision. Finally, identified constraints are georeferenced and mapped onto OpenStreetMap (OSM), supporting scalable and inclusive urban planning.
The emergence of advanced home technology and the incorporation of distributed energy resources (DERs) have markedly heightened the necessity for energy management solutions that balance technical performance with economic efficiency in smart residential microgrid (SRMG). In the absence of effective collaboration and energy interactions among smart homes, imbalances in the SRMG load profile may occur, risking violations of technical standards. This study introduces a decentralised framework for SRMG that includes diverse smart homes engaged in peer-to-peer (P2P) energy interactions. The framework is designed to minimise variations in the SRMG load profile while also reducing expenses for smart homes, all while ensuring resident comfort through P2P interactions. The home energy management (HEM) system seeks to optimise energy costs by utilising DER capabilities to facilitate P2P interactions and maintain bidirectional communication with the SRMG operator (SRMGO). Continuous data sharing between the SRMGO and HEM systems is crucial for optimising the load profile in a decentralised framework. This enables about a 4.25% reduction in load profile deviations without raising energy costs, showing that decentralised P2P energy interactions improve load management in SRMG and cost stability in smart homes. Simulation results generated using general algebraic modelling system (GAMS) software demonstrate that integrating P2P energy strategies within a decentralised framework can effectively fulfil both the technical requirements of the SRMG and the financial goals of individual smart homes.
'Zaha' is a retrieval-augmented generation question answering system integrated with The World Avatar knowledge graph, designed to support urban planning and smart city initiatives by enabling intuitive natural language queries for complex urban data. Zaha facilitates the querying of diverse domains within the urban environment, offering an accessible and effective tool for urban data analysis. By simplifying access to complex dataset, Zaha addresses a critical barrier in urban planning and management: the need for technical expertise to query data effectively. Urban data, encompassing geospatial, environmental, and regulatory information, is pivotal in informing decision-making processes. However, challenges such as data silos and the technical complexity of query tools and languages hinder the accessibility and utilisation of urban data. The integration of Zaha with The World Avatar knowledge graph further mitigates the issue of data silos by unifying urban data from diverse sources and formats into a single framework. Leveraging knowledge graph technology, Zaha facilitates efficient data retrieval based on the relationships defined between entities. By bridging the gap between data and users, Zaha empowers urban planners and other stakeholders to access and query complex urban data intuitively, enabling them to make informed decisions without requiring technical expertise.
