Forecasting power consumption is essential for intelligent power management in IoT-enabled smart environments, where heterogeneous behaviors appear from diverse building usages. University campuses are considered environments that share similarities with smart cities, making them suitable for power dynamics analysis. We build upon an IoT-based Advanced Metering Infrastructure (AMI) we deployed at our Engineering School’s Campus (INPT, Morocco), and an optimized XGBoost pipeline enhanced via Genetic Algorithms. Limited modeling granularity is addressed in heterogeneous consumption patterns. We propose and justify a cluster-aware approach partitioning data (D) into K regimes such that D=⋃c=1KCc. Each cluster is treated as a homogeneous behavioral profile and modeled using a GA-XGBoost model, enabling an intermediate granularity between global and meter-level learning. Experiments on real-world campus AMI data show that our proposed GA-XGBoost model consistently outperforms SVR and LSTM baselines across all clusters. In addition, cluster-specific models further improve performance compared to a single GA-XGBoost model trained without clustering, achieving a 48.42% improvement in MASE. Overall, beyond improving forecasting accuracy, cross-cluster generalization shows performance degradation and distributional shift when models are transferred across clusters, while residual diagnostics reveal differences in variance, temporal dependence, and non-Gaussianity.
Intelligent Transportation Systems (ITS) increasingly combine sensing, communication, computation, data platforms, and mobility services. This article presents a structured, literature-based mapping review of ITS applications from a standards-oriented perspective. The study analyzes 42 studies organized into five thematic groups and evaluated through 63 article–standard assessments using selected ITU-T Recommendations as an analytical lens. The rubric used in this work examined whether each study reported, or allowed reviewers to infer, evidence on architecture, data handling, interoperability, security and privacy, deployment assumptions, and digital-twin capabilities. Partial alignment was the most frequent outcome, accounting for 25 of 63 article–standard assessments (39.7%). At group level, satisfactory or optimal alignment occurred in 5 of 8 assessments (62.5%) in the digital-twin group and in 3 of 13 (23.1%) in the Big Data group; in the latter, 6 of 13 assessments (46.2%) showed limited or no alignment. Stronger alignment was usually found when studies described architectures, data flows, sensing mechanisms, service workflows, physical–virtual modeling, or system-management components relevant to the Recommendation, and weaker alignment when they focused mainly on algorithms, datasets, prediction accuracy, authentication, or secure dissemination without sufficient detail on interfaces, data governance, gateway roles, deployment conditions, or platform integration. The review proposes a five-dimension standards-facing reporting checklist addressing interoperability, data lifecycle and governance, security and privacy, operational readiness, and standards-facing evidence. It supports traceable reporting through explicit evidence-status categories and locations, and can be implemented as a Standards and Interoperability Reporting Statement (SIRS) for authors, reviewers, and editors. Overall, the findings show that standards-oriented assessment depends not only on technical performance but also on explicit and traceable integration evidence, while the proposed reporting profile provides a practical mechanism for making such evidence more systematically visible in future ITS studies.
Traffic signal noncompliance at urban intersections remains a road-safety and traffic-management challenge, particularly where local monitoring and empirical evidence are limited. This study characterized yellow- and red-phase stop-line crossing events at eight signalized intersections in six Ecuadorian cities and developed a preliminary site-prioritization approach. After conservative quality control, 4126 traffic-signal cycles were aggregated into 69 observation blocks. Descriptive indicators, nonparametric comparisons, Poisson and negative binomial count models, residual diagnostics, seven sensitivity analyses, and bootstrap-based prioritization were applied. The analytical sample included 38,615 vehicles and 7069 stop-line crossing events, yielding an aggregate event rate of 18.31%. Red-phase crossings represented 70.41% of recorded events and yellow-phase crossings 29.59%, with substantial variation among sampled intersections. Yellow-phase crossings were treated as protocol-defined behavioral classifications rather than event-specific legal determinations. The selected NB1 model showed no evidence of an adjusted association between off-peak and peak periods (IRR = 0.994; 95% CI 0.918–1.076), and this result remained stable across all seven sensitivity analyses. The workflow provides a reproducible basis for site-specific assessment and uncertainty-aware screening. However, the findings are not representative city-wide estimates, and the prioritization score is an exploratory within-sample tool that does not independently justify specific interventions.
Non-intrusive load monitoring (NILM) provides a cost-effective way to obtain appliance-level electricity information from aggregate smart-meter measurements and is therefore important for energy management, demand-side response, and sustainable operation in smart buildings. However, accurate appliance-level power disaggregation remains challenging because residential load signals usually involve overlapping appliance signatures, sparse activations, heterogeneous temporal patterns, and transient switching events. To address these challenges, this paper proposes a State-Event-Guided Multi-Scale Gated Network (SEMG-Net) for NILM. The proposed framework integrates a residual temporal encoder, multi-scale dilated convolutional blocks, and a state-event-guided gating mechanism within a unified multi-task learning architecture. The shared encoder extracts hierarchical temporal representations from aggregate mains windows, while task-specific branches jointly estimate appliance power, on/off state, and switching event type. The predicted state probability, three-class event probability distribution, and shared temporal representation are jointly used to construct a continuous gate that modulates the raw power estimate, thereby directly incorporating behavioral predictions into final power estimation. Experimental results on public datasets show that SEMG-Net achieves competitive overall performance, with clear advantages in power estimation, energy consistency, and state identification, particularly for appliances with complex operating stages or transient switching behavior. The ablation results further demonstrate the benefits of multi-scale feature extraction and auxiliary supervision, as well as the effectiveness of the proposed state-event-guided power modulation mechanism.
Direct evidence on curbside parking use in intermediate Latin American cities remains limited. This study characterized parking duration, purpose, turnover, accumulation, and regulatory compliance across six segment–date sessions in central Loja, Ecuador. Of 1426 observed curbside events, 1397 were retained after quality control. Analyses included descriptive and non-parametric tests, multivariable models with CR2 standard errors clustered by segment–date, and sensitivity analyses. Duration was strongly right-skewed (median 4 min; interquartile range 1–13 min; mean 36.3 min; P95 332.6 min), while passenger pick-up/drop-off accounted for 54.5% of events. Non-permitted maneuvers represented 64.0%. The four segment-sessions containing SIMERT coverage comprised 862 valid events, of which 776 occurred within marked SIMERT locations. Among the 644 marked-location events observed during payment-required hours, visible SIMERT use was recorded in 70 events (10.9%). After restricting the SIMERT component to marked locations during payment-required hours, composite non-compliance was identified in 1167 events (83.5%). During the common 06:30–18:30 comparison window, hourly turnover ranged from 0.54 to 1.96 events per legal space per hour, while cumulative space–time demand ranged from 11.6% to 154.8% of nominal legal space–time capacity. The value above 100% represents summed parking duration relative to nominal legal capacity and does not indicate simultaneous occupancy above 100%. Excluding session-boundary proxies left the median and interquartile range unchanged, although upper-tail estimates remained sensitive. Because each segment was observed on a single date, between-session differences cannot be interpreted as independent corridor effects. Natural-spline specifications provided better temporal fit than linear-hour specifications for both non-permitted maneuvers and revised composite non-compliance. These findings provide a reproducible local baseline for testing conventional and smart curb-management measures through repeated pilot studies.
With the rapid advancement of urbanization, the density and vulnerability of urban lifeline networks are increasing, and urban lifeline security risks have become a major challenge to urban public security governance. The backward governance means and fragmented governance mechanisms cannot adapt to the complex emerging urban public security risks. Digital empowerment is considered to be a new solution for the holistic governance of urban lifeline security, but related research has only focused on a single scenario, a single risk type or a single risk management link. This study shifted from a single perspective to a holistic perspective and explored how to use digital technology to develop the urban lifeline security project from three levels, that is, overall methods, key supporting technology, and governance mechanism innovation, so as to enable a holistic governance model for lifeline security. Specifically, this study constructed the main processes and methods for constructing urban lifeline security projects, the key supporting technology system for the scenario-driven urban lifeline security project, and the overall governance mechanism for the urban lifeline security project. The case from Hefei, China, further verifies the effectiveness of urban lifeline security engineering. The contribution of this study is to promote the collaborative innovation and integrated application of engineering technology and governance mechanisms in the field of holistic governance of urban lifeline security.
Advanced metering infrastructure (AMI) is the sensing backbone of the smart grid, and its reliability underpins urban energy services such as state estimation, demand response, and distributed-energy integration. When AMI uses cellular spectrum leased through a cognitive mobile virtual network operator (C-MVNO), allocating channels to data aggregation points (DAPs) each frame is difficult because three uncertainties interact: imperfect spectrum sensing, time-varying and cross-channel-correlated primary-user activity, and stochastic urban propagation. Classical Hungarian assignment is optimal per frame but blind to primary-user dynamics, while cognitive-radio heuristics ignore queue state and cross-channel structure. We propose a two-timescale hierarchy that couples these established tools in a new way: a Proximal Policy Optimization (PPO) agent decides, once per epoch, which opportunistic channels to expose, and an exact Hungarian solver performs the per-frame DAP-to-channel assignment. To our knowledge this is the first coupling of a learned cognitive layer with exact Hungarian assignment for cognitive-radio resource allocation. On a 3GPP TR 38.901-compliant simulator, PPO significantly outperforms a Bayesian-belief baseline and the Hungarian-only configuration in delivery ratio, latency, and a strict per-meter satisfaction metric, and is robust across independent seeds and sensitivity sweeps. An architectural ablation shows the DAP tier is a precondition for viability, not merely an optimization.
This paper proposes a comprehensive approach for data-driven participatory community monitoring based on “Citizen Science” (CS), ISO 37120, and artificial intelligence (AI). The design integrates AI with the CS six-stage life cycle and citizen data governance principles through an AI-CS framework, aligning with the Copenhagen Social Summit. The framework was developed for local governments in Ecuador, a country where territorial planning lacks citizen data disaggregated by territorial, sociodemographic, and contextual variables. This fact limits the capacity of local governments to make evidence-based decisions. Between October 2025 and February 2026, data from 30,253 events were collected in 22 provinces and 93 cantons of the country. The data were analyzed by means of ordinal logistic regression to identify predictors of perceived severity and by means of DBSCAN, an unsupervised machine learning clustering algorithm, to characterize territorial patterns. The results suggest that citizen perception is organized into systemic and predictable patterns when structured using ISO 37120 categories. The spatial analysis reveals heterogeneous territorial patterns with levels of urgency that differ depending on the canton and the urban–rural context. The proposed approach allows local governments to obtain disaggregated territorial data for participatory planning. Its design may be transferable to other Global South contexts facing similar data gaps and is aligned with SDGs 9, 11, 16, and 17.
Current visual perception techniques for self-driving vehicles mainly focus on the generalization across diverse scenes, and they often overlook the valuable spatial consistency present in the repetitive driving routes such as public transit lines, delivery and shuttle services. In this paper, we introduce MemGeoSeg, which is a novel multi-modal framework that enhances semantic segmentation by exploiting scene repetitions through GPS-guided spatial priors and historical memory. Our approach introduces a hierarchical GPS embedding module, which is a spatially indexed memory bank that accumulates location-specific visual knowledge and a cross-modal fusion mechanism with contrastive learning. To validate the idea of improving visual perception with repetitive driving scenarios, a new dataset, RMTD-AD, is constructed for evaluation. It contains over 13,000 annotated images across various weather and lighting conditions on repeated routes. Extensive experiments conducted on the dataset have demonstrated that MemGeoSeg significantly outperforms the state-of-the-art baseline, achieving an mIoU of 76.5% compared to SegFormer’s 71.8% (a 4.7 percentage-point improvement), with particularly strong gains in challenging scenarios like low-light and adverse weather conditions. The result shows that there are substantial benefits to incorporating geographical contexts and historical memory for location-aware perception in intelligent vehicles.
Artificial intelligence (AI), building information modeling (BIM), and digital twins are increasingly transforming construction sites into smart, data-driven environments that support safer, more efficient, and more sustainable building and urban infrastructure delivery. However, site-level spatial decision-making related to site layout optimization (SLO) remains constrained by fragmented data environments, limited interoperability, and weak integration between planning, monitoring, and adaptive decision-making. This study presents a systematic literature review of how AI, BIM, and enabling digital technologies are being applied to support smart construction site management, site-level spatial decision-making, and SLO-related applications. A Scopus-based search conducted in October 2025 identified 169 records, of which 63 studies were retained following PRISMA-guided screening. Because explicit SLO studies remain limited, the review synthesizes both directly relevant SLO studies and contextually relevant enabling studies with clear implications for smart and sustainable construction operations. The review combines bibliometric analysis, thematic content analysis, and cross-functional technology mapping to examine the intellectual structure of the field, the main operational domains addressed, and the dominant technological convergences supporting intelligent site decision-making. The findings show that the field is expanding rapidly but remains unevenly consolidated, with greater evidence concentration and practical readiness in real-time digital twin and spatial data management, automated monitoring, and proactive safety intelligence than in closed-loop logistics coordination and autonomous mobility. Across application domains, the dominant technology convergences combine machine learning and deep learning with multidimensional BIM, frequently extended through digital twins, sensors, cloud platforms, UAVs, simulation tools, and GIS-related infrastructures. The review further shows that the main barriers to deployment are not merely algorithmic, but also relate to interoperability, data quality, implementation complexity, human oversight, and limited field validation. Overall, this study provides a structured synthesis of evidence concentration, practical readiness, dominant patterns, and unresolved gaps of AI-BIM-enabled smart construction site management, and outlines directions for more interoperable, human-centered, and field-validated systems that support sustainable smart building and urban infrastructure delivery.
Smart city mobility is increasingly governed by a techno-solutionist logic that prizes data, automation, and efficiency, often at the expense of public trust, social legitimacy, and lived experience. This article argues that the fate of a mobility transition appears to depend less on the sophistication of the technology than on the pace and posture of change. Building on the CalmMobility framework and on Weiser and Brown’s concept of calm technology, it develops the idea of calm smart mobility—a human-paced, options-first approach in which innovation enters everyday life gradually and with credible alternatives already in place, so that residents are not asked to continuously adapt. The framework’s three pillars (Comprehensiveness; Pacing–Sequencing–Inclusion; Future-Readiness) are mapped onto four recurring challenges of smart mobility (Policy Layering, Affective Mismatch, Governance Silos, and the Future-Readiness Gap) and then used as a descriptive analytical lens to characterize seven documented implementations across economic, spatial, mass-transit, service, and platform interventions and four world regions: the Stockholm congestion charge, the London ULEZ expansion, the Barcelona superblocks, Bogotá’s TransMilenio bus rapid transit and Ciclovía, Seoul’s Cheonggyecheon restoration and bus reform, Helsinki’s Whim Mobility-as-a-Service, and Sidewalk Toronto. Presented through a comparison table, a positioning map, and adoption trajectories rather than rankings, the characterization suggests that the provision of alternatives, the sequencing and pace of change, and the genuineness of co-creation are more closely associated with smooth adoption than the type of instrument deployed. The article is conceptual and framework-building. The cases illustrate and probe the framework instead of validating it, and a testable central hypothesis is specified for future empirical work. Calm smart mobility is offered as a transferable, citizen-centred logic for guiding smart city mobility transitions at a human pace.
Smart cities increasingly rely on urban digital systems deployed across domains such as mobility, public safety, surveillance, and governance, involving large-scale collection and processing of sensitive data. These systems raise significant cybersecurity and privacy challenges, shaped by European regulatory frameworks that influence how data are collected, secured, shared, and governed within urban environments. While existing research has examined legal and regulatory aspects alongside technical cybersecurity solutions, these areas are often addressed in isolation, limiting insight into how regulatory requirements translate into concrete implementations. This paper presents a comprehensive review of regulatory-driven cybersecurity approaches for smart cities. It maps the literature across major application domains and analyses how regulatory objectives are reflected in technical, organisational, and operational measures, as well as in implemented solutions. By jointly examining legal and technical perspectives, the review links regulatory compliance requirements with concrete security practices and system-level design choices. Based on this analysis, the paper proposes a structured classification of regulatory-driven smart city approaches and identifies key trends, gaps, and challenges in the literature. The findings provide a foundation for future research on regulatory-driven cybersecurity and privacy protection in smart systems.
Artificial Intelligence (AI), particularly Machine Learning (ML) and Deep Learning (DL), is transforming the modeling of complex spatiotemporal urban processes such as urban growth, sprawl, shrinkage, redevelopment, and Land Use/Land Cover Change (LULCC). However, despite rapid methodological innovation, applications remain fragmented, and there is limited synthesis of how AI-based models complement, extend, or supersede conventional approaches. This study addresses this gap through a systematic review of 6356 records, from which 120 articles were selected for detailed analysis. It investigates: (i) how ML/DL techniques are embedded within spatiotemporal modeling frameworks; (ii) their use in simulating urbanization dynamics and land-use (LU) transitions; (iii) methodological and performance gains relative to traditional statistical and rule-based models; and (iv) emerging research frontiers and limitations. The review shows that LULCC dominates current applications, with Artificial Neural Networks (ANNs) as the most prevalent ML method, increasingly complemented by DL architectures. Across cases, AI is primarily used to learn non-linear transition dynamics, represent spatial and temporal dependencies, identify influential drivers, and improve classification performance and computational efficiency. Building on these insights, the paper synthesizes the roles of AI in spatiotemporal urban modeling and outlines forward-looking research directions to support more robust, transparent, and policy-relevant applications for urban sustainability.
Urban flexibility research is expanding across buildings, electric vehicles (EVs), distributed energy resources (DERs), storage, positive energy districts (PEDs), digital twins, and interoperability platforms. These strands are often reviewed separately, although urban distribution operators must manage their combined impacts on the same feeders. This paper presents a PRISMA 2020-aligned systematic review with evidence mapping and narrative synthesis of feeder-aware coordination in smart-city electricity systems. Searches of Scopus, Web of Science, IEEE Xplore, ScienceDirect, and citation chasing identified 312 records; 127 studies were included after screening and eligibility assessment, 101 entered the quantitative mapping sample, and 31 formed the deep-synthesis anchor core. Sparse contingency tables were analyzed with Monte-Carlo permutation chi-square tests and bootstrap confidence intervals for Cramér’s V, while ordinal variables were summarized with medians and interquartile ranges. Explicit feeder grounding was concentrated in grid-oriented and EV-oriented studies, whereas many AI/digital-twin and interoperability studies were less often validated against distribution-network operation. Economic and peak-flexibility indicators were reported far more often than interoperability, cybersecurity, or validation-maturity indicators in the anchor core. The synthesis also showed that deployment-oriented work depends on clearer treatment of standards, co-simulation workflows, regulatory instruments, and stakeholder roles. The evidence base is heterogeneous, English-only, and single-coded, so the quantitative results are descriptive rather than population-level. The review contributes a transparent three-layer corpus design (127 included/101 mapped/31 anchor), a domain-specific specialization of SGAM/IEEE 2030 for urban feeder orchestration, an operational digital-twin definition and validation ladder, a retrofittable benchmarking framework, and a practical roadmap for DSOs, municipalities, aggregators, EV operators, building managers, and ICT providers.
This work explores how graph theory and graph neural networks can support the strategic planning of rail network expansions using only publicly available city data, applied to the São Paulo Metropolitan Region. The methodology consolidates information from multiple public sources, develops a catchment-area formula to estimate potential passenger demand, applies Random Forest to identify the most relevant demographic features, and implements a GraphSAGE model that derives predictive capability from network topology together with socioeconomic features and origin–destination trips. The demand approximation was checked against observed station boardings, with predicted and observed rankings in agreement. The GraphSAGE model achieved an R2 of 0.874 ± 0.042 when predicting the proxy demand indicator, with minimal overfitting, outperforming the Random Forest baseline and achieving accuracy comparable to an XGBoost baseline while overfitting substantially less; this performance remained stable under spatial cross-validation. The model is computationally efficient and requires no rail-system-specific information beyond topology, making it suitable for the fast, low-cost comparison of expansion proposals rather than as a replacement for detailed transport demand models. It was used to evaluate eleven real projects and proposals for the São Paulo Metropolitan Region. Employment, residences, and destinations where people go to eat together represent about 65% of the model’s predictive capacity.
The ambitious roadmap for a sustainable transport system adopted by the European Commission (EC) by 2050 includes the deployment of an extensive Electric Vehicle Charging Stations (EVCSs) infrastructure, which introduces significant challenges for distribution power grids. High power demand, particularly from fast-charging systems, may lead to network overloading and voltage unbalance. In addition, recent measurement campaigns highlight substantial changes in grid impedance and the emergence of resonance phenomena, together with the injection and propagation of high-frequency conducted disturbances. These effects extend over a wide frequency range, up to several hundreds of kHz, causing degradation, aging and malfunction of network assets, in particular Power Line Communications. This paper provides a comprehensive and updated review of the impact of EVCSs on electrical grids, covering power flow, power quality, stability, and impedance-related interactions. Particular attention is given to the role of power-electronic converters, high-frequency emissions, and the associated challenges in measurement and standardization. The analysis highlights that EVCS integration fundamentally alters the nature of electrical loads, requiring new approaches for grid planning, monitoring, and regulation. The study identifies key research gaps and outlines future directions to ensure the reliable and sustainable integration of electromobility into modern power systems.
Urban traffic anomaly detection is essential for intelligent transportation systems, particularly in smart city environments where fast identification of abnormal events can improve road safety and traffic management. This work proposes a novel ULSTM-driven architecture that explicitly models temporal dependencies across consecutive traffic frames to achieve more stable and temporally coherent reconstructions. The proposed framework leverages sequential spatio-temporal representations to improve the distinction between normal traffic patterns and anomalous events. To further enhance reliability, we introduce a Hybrid Weighted Fusion strategy that synergistically combines structural, perceptual and pixel-wise metrics. The framework’s parameters are optimized using a Discrete Dirichlet Sampling approach, achieving a peak F1 Score of 70.28%. Evaluations were conducted on a manually curated traffic anomaly dataset with frame-level annotations. Experimental results demonstrate that the ULSTM framework significantly outperforms frame-independent generative models by suppressing high-frequency reconstruction noise, providing a robust solution for real-world smart city deployments. While highly effective in complex scenarios, the proposed framework is strictly applicable to highly dynamic traffic environments with active motion, as static background ensembles can degrade performance.
Firefighters operate in high-risk, rapidly evolving environments where exposure to extreme heat, toxic gases, and physiological stress significantly increases the likelihood of injury and fatality. This study systematically maps the emerging research landscape of real-time artificial intelligence (AI)-driven digital twins for environmental and physiological risk prediction in firefighting contexts. A combined bibliometric and qualitative content analysis was conducted using peer-reviewed literature retrieved from the Web of Science database (2010–2025). Bibliometric techniques were used to identify publication trends and thematic clusters, while content analysis examined the integration of sensing technologies, AI models, and digital twin architectures. The results reveal four dominant technological domains shaping the field: AI-enabled fire risk modeling, sensor data acquisition systems, IoT-based digital infrastructures, and predictive analytics for disaster simulation. Sensing technologies such as temperature, gas, particulate matter, thermal imaging, heart rate, and blood oxygen monitoring form the foundational data layer, while machine learning and deep learning models enable real-time hazard prediction and situational awareness. Digital twin architectures serve as the integration layer, fusing multi-source data and supporting simulation-based decision-making. Despite rapid advancements, key gaps persist, including limited integration of environmental and physiological data, insufficient predictive capabilities, a lack of standardized architectures, and minimal development of human-centered decision-support systems. This study provides a structured synthesis of current technologies and identifies future research directions toward integrated, explainable, and real-time digital twin systems to enhance firefighter safety and operational resilience.
Urban land use intensity (U-LUI) is a widely used term for describing urban development processes, yet its conceptualisation and measurement remain inconsistent. Existing approaches focus on isolated dimensions, such as structural density, functional activity, and socio-economic indicators, resulting in limited comparability and weak integration across scales and data sources. This paper reviews and synthesises current approaches to U-LUI with a focus on remote sensing (RS), in situ data and emerging urban data sources. It analyses definitions, related concepts of urban intensity and existing monitoring frameworks at national, European and global levels, and compares methodological approaches for observing U-LUI. Based on this synthesis, U-LUI is defined as a continuous, multidimensional and spatio-temporally dynamic property of urban systems that reflects the intensity of anthropogenic use. To operationalise this concept, the paper develops an integrative, trait-based framework comprising six indicator families: traits, genesis, structure, taxonomy, function and socio-economics. The proposed framework is illustrated and supported through the synthesis of existing RS approaches, urban monitoring concepts and representative examples from the literature, demonstrating its potential for consistent and scalable U-LUI assessment. These dimensions link physically observable characteristics with functional and contextual aspects of urban systems and provide a basis for more consistent quantification and comparison. The results highlight key challenges for U-LUI monitoring, including limited conceptual harmonisation, incomplete integration of dimensions and the need for improved data integration. The proposed framework supports more coherent and scalable assessments of U-LUI in research, monitoring and planning contexts.
Artificial intelligence is reshaping employment in smart cities, yet centralized hiring platforms can deepen exclusion for persons with disabilities through privacy risk, biased models, weak multilingual support, and limited accommodation awareness. Because disability-related records are highly sensitive, no single institution holds enough representative data to train fair models, and centralizing such data is rarely permissible across borders. We propose FedAgent-Chain, a framework that integrates federated learning, blockchain-based auditability, multilingual processing, rule-based agentic services, and human-in-the-loop governance, extended with an education-to-employment module that builds individualized, accessible job-readiness pathways. Institutions across Saudi Arabia, the United States, China, and Europe train shared models without exchanging raw data. In a prototype evaluation on synthetic records over five seeds, the framework reached a mean F1 of 0.7207 (95% CI: [0.6506, 0.7909]), comparable to a centralized logistic-regression baseline while preserving data locality, with a formal (ε=3.2,δ=10−5) differential-privacy guarantee after 20 rounds. Multi-dimensional fairness regularization lowered disability-category and work-mode disparity by 32.3% and 40.3% relative to local-only training. We report the fairness behavior transparently, including a case where the penalty does not outperform standard FedAvg on disability-category disparity, and we position cross-institutional integration with accountable governance, rather than raw metric superiority, as the central contribution.
