The spatiotemporal nonstationarity of shared bicycle usage, a sustainable and eco-friendly mode of transportation, is believed to be influenced by the built environment. However, the specific spatial and temporal impacts of built environment factors on shared bicycle trips are not yet fully understood. This study investigates the relationship between the built environment and shared bicycle ridership in Shenzhen, a city where the distribution of shared bicycles is relatively dense, by utilizing multisource urban big data. Key independent variables were selected based on the “5Ds” dimensions of the built environment, and the performance of two models—Geographically Weighted Regression (GWR) and Geographically and Temporally Weighted Regression (GTWR)—were compared. The analysis evaluates the impact of the built environment on the density of shared bicycle ridership, incorporating both spatial and temporal dimensions. The results of the study found that the GTWR model used in this paper can effectively explain the spatio-temporal heterogeneity of built environment-related variables on shared bicycle trips with high goodness of fit. And the regression fit coefficients of the model show that the effects of different built environment indicators on the density of shared bicycle ridership are significantly different in both time and space. Among them, road network density, catering POI density, traffic POI density and POI diversity have a facilitating effect on shared bicycle travels, particularly during peak hours on weekdays and in central urban areas. Shopping POI density shows different effects on shared bike use in different times and spaces. While the distance from the city center and the nearest distance to the bus station have a suppressive effect on shared bicycle use, they show opposite degrees of influence in the spatial distribution. The results can provide more precise guidance for future rational transportation strategies or sustainable urban planning.

Accurately predicting commuting flows is crucial for sustainable urban planning and preventing disease spread due to human mobility. While recent advancements have produced effective models for predicting these recurrent flows, the existing methods rely on datasets exclusive to a few study areas, limiting the transferability to other locations. This research broadens the utility of state-of-the-art commuting flow prediction models with globally available OpenStreetMap data while achieving prediction accuracy comparable to location-specific and proprietary data. We show that the types of buildings, residential and non-residential, are a strong indicator for predicting commuting flows. Consistent with theoretical and analytical models, our experiments indicate that building types, distance, and population are the determining characteristics for mobility related to commuting. Our experiments show that predicted flows closely match ground truth flows. Our work enables accurate flow prediction using building types to support applications such as urban planning and epidemiology.

Nowadays, number of private cars is increasing rapidly. Traffic congestion becomes a serious problem in urban region. If traffic congestion can be predicted before it happens, it will be helpful for improving traffic condition. So many traffic congestion prediction methods have been proposed. Almost all these methods are based on traffic flow prediction algorithm. In these methods, historical traffic flow data is used while performing prediction. Obviously, information of sudden accidents like traffic accidents, road damage and bad weather that happened recently may be not contained in historical traffic flow data. But performance of traffic flow prediction algorithms will be affected by these factors. In this situation, performance of traffic congestion prediction method based on traffic flow prediction result will be affected as well. To solve the problem, a new traffic congestion prediction method based on trajectory mining algorithm is proposed in this paper. In this method, traffic controllers can set a threshold for each road according to the current situation of the road. The threshold represents the vehicle number that can be carried by the corresponding road in a short period. Besides, for each road, the proposed method tries to count the number of vehicles that will pass through the specific road at next time step by predicting next location for all the running vehicles based on their trajectories. If the vehicle number of a road surpasses the threshold of this road, it will be predicted as congested road. Otherwise, it will be predicted as non-congested road.

The Spatial Data Lab (SDL) project is a collaborative initiative by the Center for Geographic Analysis at Harvard University, KNIME, Future Data Lab, China Data Institute, and George Mason University. Co-sponsored by the NSF IUCRC Spatiotemporal Innovation Center, SDL aims to advance applied research in spatiotemporal studies across various domains such as business, environment, health, mobility, and more. The project focuses on developing an open-source infrastructure for data linkage, analysis, and collaboration. Key objectives include building spatiotemporal data services, a reproducible, replicable, and expandable (RRE) platform, and workflow-driven data analysis tools to support research case studies. Additionally, SDL promotes spatiotemporal data science training, cross-party collaboration, and the creation of geospatial tools that foster inclusivity, transparency, and ethical practices. Guided by an academic advisory committee of world-renowned scholars, the project is laying the foundation for a more open, effective, and robust scientific enterprise.

With the orderly progression of urban renewal in China, social capital, as an important factor in analyzing the relationship between stakeholders, has once again become a key topic in urban planning research. Green spaces have proved to have a more notable impact on social capital than other built environment factors. In light of this, the present study is based on the data of 1,282 residents’ questionnaire surveys conducted in Nanjing in 2022 and extracts spatial characteristic information of green space through multisource big data such as remote sensing data, street-view data, and points of interest, to investigate the influence mechanism of physical and perceived attributes of urban green space on community social capital. The study found that there are differences in the influence mechanisms and dimensions of these dual attributes of green space on social capital. The direct effect of perceptual attributes is more substantial, while physical attributes almost exclusively affect social capital indirectly through the perception of green space. Notably, among the physical attributes, only the total coverage of neighborhood vegetation has a considerable direct effect on neighborhood relations, whereas community sentiment, a willingness to participate, and larger and more aggregated green spaces do not enhance community social capital effectively. Lastly, community social capital is affected substantially by exogenous variables of socioeconomic attributes, and there is group differentiation. Results reveal the direction of renewal and optimization of urban green spaces from the perspective of promoting social capital, which provides a reference for the synergistic and high-quality development of the community’s physical and nonphysical environments.

Human dynamics research has undergone a significant transformation over the past decade, driven by interdisciplinary collaboration and technological innovation. This opinion paper examines the evolution of the field in the past ten years, focusing on its integration of GIScience (Geographic Information Science), social science, and public health to tackle spatial and societal challenges such as urban sustainability, disaster response, and epidemics. Key advancements include the adoption of living structure theory, which redefines space as a dynamic and interconnected entity linked to human well-being and ecological sustainability, and the application of cutting-edge technologies like GeoAI (Geospatial Artificial Intelligence) and digital twins for adaptive modeling and informed decision-making. Despite these advancements, challenges persist, including incomplete data, mismatched scales, and barriers to equitable access to geospatial information. Addressing these issues necessitates innovative approaches such as multiscale modeling, open data platforms, and inclusive methodologies. Increased funding opportunities offer pathways for accelerating translational research. By integrating advanced theories, user-centered technologies, and collaborative frameworks, human dynamics research is poised to transform urban systems into sustainable, resilient, and equitable environments. This paradigm shift underscores the importance of ethical considerations and inclusivity, offering a holistic approach that aligns with human and ecological needs.

The original online version of this article was revised: The authors Ming-Hsiang Tsou and Di Zhu are switched in their affiliations.

A correction to this article is available online at https://doi.org/10.1007/s43762-025-00202-z.

The evidence on the relationship between built environment factors and obesity in primary school children is limited, and this study is the first to investigate this relationship in Iran. This study utilizes Geographical Information Systems (GIS) techniques to assess built environment indices for geographical addresses based on the street network. A school-based survey was conducted in ten neighborhoods in Tehran from January to April 2019, collecting socio-demographic information and home addresses from 2,677 primary school children (6–13 years). School nutrition experts measured children's height and weight, and their obesity status was calculated based on the BMI z-score adjusted for age and gender. Logistic regression analysis showed that higher accessibility to parks within 2 km was associated with lower odds of obesity, even after adjusting for age, gender, family income, and parental educational level in the model (OR = 0.919, 95% CI = 0.848–0.996). Living in an area less than 400 m from a park was also associated with lower odds of obesity (OR = 0.811, 95% CI = 0.665–0.989). Access to sports facilities and the percentage of major streets were inversely associated with childhood obesity (highest vs. lowest tertile OR = 0.766; 95% CI = 0.597, 0.985 and OR = 0.739, 95% CI = 0.582, 0.938 respectively). However, no significant relationships were identified for residential density, intersection density, land-use diversity, and the effective walkable area index. Similar to findings from other international studies, these results suggest that addressing spatial disparities in access to parks and sports facilities as an amenable environmental factor is important for reducing children's obesity. This information is valuable for creating local policies and intervention programs. Further investigations with a longitudinal design may provide a better understanding of these relationships.

The use of ride-hailing services, online shopping, and telecommuting are behaviors which have recently increased dramatically in popularity, due in part to technological advancement and global events such as the Covid-19 pandemic. In theory, these behaviors may have the potential to shift people towards more sustainable travel. This study aims to explore the influences of ride-hailing, online shopping, and telecommuting on household vehicle miles traveled (VMT) and walking trip generation using the 2017 and 2022 U.S. National Household Travel Surveys (NHTS). Results reveal that the frequencies of all three activities increased between 2017 and 2022, online shopping and telecommuting showing positive correlations with VMT generation, and higher mean VMT associated with all three activities in 2017 and with online shopping and telecommuting in 2022. Regression models further indicate that telecommuting is most strongly associated with more sustainable travel, with seven of the eight models estimated indicating lower VMT generation and more walking trips associated with telecommuting. Ride-hailing service usage was also associated with lower VMT and more walking trips in six models. The results for online shopping are mixed, with our models showing that online shopping leads to more walking trips, but also higher VMT. The results of this study indicate that ride-hailing services and telecommuting may play an important role in shifts towards more sustainable travel behavior. Suggestions are presented for maximizing shifts to sustainable travel modes and minimizing potential inequitable effects, including designing for increased walkability, particularly in predominately minority areas, and the promotion of transit-oriented development.

Digital twins are enjoying widespread and growing success in both theoretical and practical applications. A recent development that is gaining increasing traction is the application of digital twins to cities. The aim of this article is to discuss whether there are inherent limitations in this case. At present, the scientific literature on urban digital twins is dominated by “technical” approaches. Critical investigation of digital twins – especially from a philosophical perspective – is still at its beginnings. This article aims to contribute to this line of inquiry. It is mainly theoretical and analytical. On the basis of a specific conceptual framework, it examines digital twins and their applications in urban contexts. It starts by distinguishing among simple, complicated and complex systems, and reaches the conclusion that, while using digital twins is generally appropriate (and often helpful) in the first two of these systems, there are some structural limitations on their use in the case of complex systems. In the latter case, inherent limitations depend on certain distinctive aspects of complex systems, such as their emergent and unpredictable nature, and the role played in this regard by “dispersed knowledge” (that is, is a form of diffused practical knowledge that is crucial for the functioning of large urban systems but that cannot be collected and re-unified because, as a coherent and integrated whole, it does not and cannot exist anywhere).

Urban areas globally have become home to over half of the world's population, leading to the intensification of the urban heat island (UHI) effect, where cities experience higher temperatures than their rural counterparts. The current study develops a new model predicting UHI intensity for 216 cities across all climate zones for both the Global North and Global South using machine learning techniques, focusing on the years 2019 to 2023. Utilising a novel dataset, integrating climate, economic, population, and land use data from 216 cities worldwide, the model, trained using Support Vector Regression (SVR), demonstrates a mean absolute error (MAE) of 0.86 °C. Results reveal that wind speed significantly mitigates UHI intensity, while cities in temperate climates exhibit more pronounced UHI effects compared to those located within tropical climbs. Additionally, results show the crucial role of coastal proximity in reducing UHI intensity and find no significant differences in UHI intensity between cities in the Global North and Global South. Findings offer important empirical actionable insights alongside a robust tool for urban planners and policymakers to measure, map, and monitor the UHI effect, contributing to the development of more liveable and sustainable urban environments.

The digital economy drives economic growth and regional competitiveness. Understanding the evolution of county-level digital economies is essential for regional economic transformation, upgrading, and long-term development. Traditional assessment methodologies have several shortcomings for representing the county digital economy, especially data availability and reliability. In this paper, we develop a multi-scale analytic framework using complex network indicators including average clustering coefficient,

Urban Digital Twin (UDT) technology is increasingly recognised as a promising tool for designing and developing sustainable, resilient urban environments. Nonetheless, the current literature lacks a comprehensive understanding of UDTs’ current applications in the built environment. Therefore, this study addresses the identified gap by analysing scholarly literature and industry reports connected to UDT implementations. The results of scientometric analysis revealed five key research fields including: (i) UDT for urban monitoring and controlling, (ii) UDT for smart urban planning, (iii) UDT for environmental management, (iv) UDT for decision-making, and (v) UDT for smart and sustainable cities. Further, this study analysed 10 industry reports on UDT technology to identify practical insights and evaluate industry-driven approaches for implementing UDT solutions in urban environments. Despite promising progress, the findings indicate the absence of a clear, structured process to facilitate consistent implementation, scalability, and interoperability in UDT technology. This further highlights the need for globally recognised guidelines and well-defined KPIs to fully realise its potential in urban environments. The study also presents a new classification model developed from analysing the research flow to elaborate on the main outcomes from five clusters towards UDT pathways. The new proposed model reintroduces the structure of UDT literature with a new flow to interpret and correlate the content identified in previous studies. Based on these insights, the study offers recommendations to support the advancement of UDT technology for building resilient, sustainable cities.

As urbanization continues to accelerate, there is a growing need for the analysis of urban spatial perception attributes and scene elements. In response, the research proposed a multi-scale perception network-based model for extracting urban scene elements and an attention-enhanced segmentation network-based model for analyzing urban spatial scene structures. The urban scene feature extraction model incorporated Siamese convolutional neural networks and convolutional block attention to achieve multi-scale perception extraction. The urban spatial scene structure analysis model combined a dynamic attention module with an encoder-decoder architecture to enhance the accuracy of scene element segmentation. During testing of the urban scene feature extraction model at a resolution of 768, its classification accuracy and cross entropy were 95.4% and 0.065, respectively. The model's average ranking accuracy for beauty, comfort, and cleanliness was 92.5%, 91.8%, and 93.2%. In testing the urban spatial scene structure analysis model, the boundary intersection to union ratio and boundary F1 score were 81.2% and 82.1%, respectively, with a boundary complexity of 0.6. The results demonstrated that the proposed method excelled in tasks such as perceptual attribute classification and scene element parsing, effectively addressing complex and diverse urban spatial features.

Effective landscape design, which optimizes solar irradiation and absorbed heat to reduce mean radiant temperature (MRT), is critically important for enhancing outdoor thermal comfort, particularly given the ongoing global decline in vegetated areas. Using Cairo, Egypt as a case study, it examines how variations in design layouts and surface materials influence reflected solar energy and, consequently, the urban thermal climate. The research introduces a novel approach by integrating generative design and parametric modeling to optimize urban park microclimates, offering a structured methodology for sustainable and climate-resilient urban spaces. A parametric modeling approach was employed to test various landscape configurations, adjusting paving size, material distribution, Rotation angle for Landscape, and tree placement. Over 1,500 design cases were simulated and analyzed using the Climate Studio plugin for Grasshopper 3D. Through generative design algorithms, an optimized framework was developed to identify effective strategies for urban cooling. Findings indicate that smaller, scattered hardscape patterns with maximum 6% divided ratio, combined with light and dark surfaces, wood, and softscape areas, significantly reduce reflected solar energy. Tree placement over light-colored hardscapes proved effective in lowering solar reflection, while optimal hardscape tile rotations towards to northern orientation. Multi-variable scenarios optimization incorporating trees, water surfaces, and reflective materials achieved up to a 44% reduction in solar irradiation. These results highlight the importance of balancing hardscape and softscape areas, integrating vegetation and water features, and utilizing light-colored materials in dispersed patterns. The study provides actionable insights for urban planners and landscape architects to design sustainable, climate-adaptive cities.

During the pandemic, urban parks have played an increasingly important role in promoting urban sustainability by providing outdoor leisure activities. While some parks have served as havens from social isolation, promoting outdoor activity and physical and mental health, others have experienced a decline in visitors. This study aims to classify urban parks based on their visiting patterns during the pandemic and identify the locational factors and design elements that contribute to their typology. By analyzing location-based big data from 425 parks in Seoul, South Korea, we utilized a multinomial logit model and K-shape clustering to explore how park characteristics are linked to clusters with different visiting patterns. The findings reveal that Children's Parks, District Parks, and parks with sports facilities and appropriate sizes tend to be classified as a type with significantly increased visits during the pandemic era.

Exploring the impacts of population place visitation on crime patterns is crucial for understanding crime mechanisms and optimising resource allocation in crime prevention. While recent studies have broadly examined dynamic population activities at specific places from geo big data, limited crime-related studies have utilised this measurement to disentangle the impact of specific place visitation on urban crime patterns. This study aims to investigate the impact of population activities at different urban functional places on theft levels across different urban areas and distinctive social changing contexts. We utilised geo big data (mobile phone GPS trajectory records) collected from millions of anonymous users to measure footfalls (counts of visitations) attached to place types on weekdays and weekends. An explainable machine learning approach was applied to analyse the impacts of place visitations on theft levels: the ‘XGBoost’ algorithm trained a high-performance regression model and ‘SHapley Additive exPlanations’ (SHAP) values were measured to identify the contributions of different visitation variables to theft levels at specific spatial and temporal scales. Using the police records and geo big data in Greater London from 2020 to 2021, the optimised model revealed that visitation to ‘Accommodation, eating and drinking’ services during weekdays had the most significant impact compared to 17 other types of place visitations. Further, the influence of place visitations on theft varied across different local urban areas corresponding with changes in social restrictions during the pandemic. Specifically, the urban areas where theft was most impacted by visitation at specific types of places (e.g., accommodation, eating and drinking services) shifted to outer London during the first national lockdown compared to normal times. The findings provide further evidence from direct micro-level analysis and contribute to tailoring policing strategies in places with different contexts and urban visitation patterns.

The rapid evolution of Artificial Intelligence (AI) has ushered in a transformative era for urban studies, moving beyond traditional analytical methods to advanced Deep Learning architectures, with Transformers model in the spotlight. Yet, unlike bioinformatics, which has successfully utilised AI to decode static biological systems, or cheminformatics, which optimises chemical synthesis, urban informatics grappled with human-centric complexity that encompass subjective perceptions, socio-political dynamics, and multifaceted challenges that defy deterministic solutions. To avoid techno-solutionist pitfalls, we convened an interdisciplinary group of scholars to explore AI-powered urban informatics and proposed a Human-AI Symbiosis framework to foster sustainable cities and advance urban research. This Opinion paper synthesises insights into four key research directions, focusing on the evolving landscape of urban informatics and its potential to drive innovation in sustainable cities, policy-making, and societal development.

Urban transportation networks are recognized for their pivotal role in forecasting city indicators and facilitating efficient planning and management. However, despite the increase of methodologies and models harnessing machine learning advancement to forecast these values, challenges persist in scenarios where direct demographic or economic data are limited or unavailable. In this paper, we propose an approach to infer socioeconomic information in an urban context without relying on traditional, official data sources, but rather focusing on publicly available data relating to the digital footprints of the cities’ inhabitants. We leverage Graph Neural Network (GNN) models to capture the spatial relationships inherent in network data while integrating perceptual features extracted from images to enhance predictive accuracy. Our results demonstrate that the combination of these data sources enables a GNN to achieve robust performance in predicting socioeconomic indicators, particularly in settings where traditional demographic and economic data may be sparse or unavailable. Through our analysis, we show that while perceptual features alone offer substantial predictive power, the inclusion of map topology through GNN models provides crucial context, leading to better generalization and more reliable predictions across different urban areas.

Local Climate Zones (LCZs) exhibit distinct thermal environments, yet their spatial configuration also governs local heat dynamics. This study examines the relationship between LCZ landscape patterns and Land Surface Temperature (LST) in Bhopal, India. The city contains nine LCZ classes, of which compact low-rise (LCZ-3), open low-rise (LCZ-6), and sparsely built (LCZ-9) dominate the urban fabric and show significant thermal variation (ANOVA: F (2,13) = 19.306, p < 0.005). Spatial indices of urban morphology, including patch and edge densities, were strongly correlated with LST across 16 LCZs (r = 0.527 and 0.800, p < 0.05). LCZ-3, marked by high patch and edge densities, recorded the maximum surface temperature (42.9 °C), confirming the thermal burden of dense built-up forms. In contrast, LCZ-9 exhibited irregular morphology, where shape indices (SHAPE_MN and SHAPE_AM) showed significant negative correlations with LST (r = –0.849 and –0.739, p < 0.05). Further validation across three urban form typologies of Bhopal’s urban developed area revealed significant differences in thermal response among spatial typologies (ANOVA: F = 4.17, p = 0.002). Overall, compact and dense morphologies were associated with elevated LST, while irregular and less compact forms displayed lower thermal intensity. These findings demonstrate the critical influence of spatial patterns on microclimates and highlight the role of climate-sensitive urban planning and built-form regulations in mitigating heat stress.

To address the challenges of global urbanization and housing shortages, implementing practical densification approaches often necessitates tailoring solutions based on the local context and prevailing housing typologies. However, such expansion strategies have been limited to unidirectional stacking or single-direction extensions while also heavily relying on designers' previous experience and subjective judgment. Therefore, this study proposes a novel machine learning (ML)-based framework for generating multi-directional house extension options, enabling efficient and contextually appropriate residential densification. Unlike existing approaches limited to unidirectional expansion and subjective designer input, our framework automatically identifies suitable land surfaces, conducts voxel-based generation of extensions, and incorporates customizable, structurally valid prefabricated components. A case study of Walthamstow, a neighborhood in North London, UK, demonstrates the framework's potential for significant residential densification. Key findings reveal that our proposed data-driven approach can generate scalable densification solutions tailored to diverse residential building types and neighborhoods, offering a promising strategy to reduce urban sprawl, alleviate the housing crisis, and minimize environmental impact through efficient, automated, and contextually sensitive design. This ML-based framework significantly advances automated densification strategies, providing a practical tool for sustainable urban development.

This paper discusses the development and application of a digital twin (DT) for urban resilience, focusing on an integrated platform for real-time fire and smoke. The proposed platform, FireCom, adapts DT concepts for the unique challenges of urban fire management, which differ significantly from regional wildfire systems. Through an exploratory case study in Austin, Texas, in the United States, this research bridges the theoretical foundations of 3D DT with their practical application in fire and smoke management. By fusing diverse data sources, ranging from air quality sensors and meteorological data to 3D urban infrastructure, FireCom supports both emergency response and public awareness through a publicly accessible dashboard. Unlike platforms developed primarily for wildland fire applications, FireCom is specifically designed to account for urban complexities such as building canyon effects on smoke dispersion and the heightened exposure risks associated with dense populations. This study contributes a scalable, replicable framework for municipalities seeking data-driven tools for proactive disaster management, with implications for broader climate resilience planning in urban areas.

Street-level visual experiences are underrepresented in top-down spatial datasets such as remote sensing and spatial footprints, which predominantly capture configurations from an overhead perspective. This study develops a framework to model and map urban visual dominance, defined through seven typologies based on Greenness, Openness, and Enclosure. A total of 12,631 Google Street View panoramas were semantically segmented with a pretrained ADE20K deep learning model to extract proportions of trees, buildings, and sky. These proportions were aggregated into 50 m hexagonal grids and classified into visual dominance classes through rule-based logic. To predict these classes beyond street-view coverage, three scenarios of spatial predictors (remote sensing indices, building footprints, and their combination) were evaluated using five machine learning algorithms. Logistic Regression with combined predictors performed best, achieving an accuracy of 0.503 and an AUC-ROC up to 0.85 for the Greenness class. External validation against GHSL settlement data across six urban sites showed soft accuracy scores ranging from 22.33% to 67.23%, with better performance in structured environments than in fragmented residential settings. These findings highlight both the promise and limitations of generalizing street-view visual information from two-dimensional spatial features, offering a scalable approach to bridge the spatial coverage gap and support more human-centered urban landscape analysis.

Advances in Machine learning open new frontiers in the systematic analysis of urban form. The study presents a scalable and interpretable framework that derives an urban-form typology by performing unsupervised clustering of 17 multi-scale morphological indicators encoded at the cadastral plot scale. The method adds positional information with the Getis-Ord Gi* spatial autocorrelation metric to encourage spatially homogeneous clusters. The study employs a combination of UMAP for non-linear dimensionality reduction and BIRCH for scalable clustering. Caveats of using the plot as a spatial unit are mitigated via filtering, tessellation and buffering. Applied to the metropolitan area of Thessaloniki, Greece, the framework identifies 14 urban form types organized into five families with similar characteristics. The resulting typology reveals, in a Conzenian fashion, patterns of urban development rooted in the city’s modern history. Results are validated quantitatively with performance metrics and qualitatively using aerial imagery and established knowledge of Thessaloniki’s planning and evolution.

Shared bike services are rapidly expanding across the globe due to their potential to alleviate urban transportation congestion, offer environmental benefits, and address first- and last-mile connectivity. The city of Austin, Texas, is actively working to scale up its MetroBike system by increasing the number of available bicycles and expanding bike lanes. While previous studies have primarily focused on the accessibility or community detection separately, relatively little attention has been given to comprehensive usage patterns, spatial accessibility, and the factors that influence them. Understanding these components is crucial for the efficient and sustainable expansion of shared micromobility services. This study investigates the MetroBike system in Austin through a multi-method approach. First, a community detection analysis is used to identify spatial clusters of bike-sharing activity, revealing variations in urban structure and usage across space and time. Second, we employ a two-step floating catchment area (2SFCA) method to evaluate accessibility to bike kiosks at both the census block group and community levels. Finally, we examine the socio-demographic factors that influence accessibility based on a geographically weighted regression (GWR). The findings provide critical insights into spatial disparities and usage trends in Austin's bike-share system, offering data-driven guidance for the equitable and strategic expansion of MetroBike. This research contributes to the broader understanding of micromobility accessibility and supports urban policy aimed at sustainable transportation planning.

Road traffic accidents (RTAs) cause approximately 1.35 million deaths and 50 million injuries annually, disproportionately affecting people aged 5–29 years. The objective of this review was to synthesize how Geographic Information Systems (GIS) support RTA analysis and road safety audits. Relevant articles were searched in different electronic databases such as Scopus, Web of Science, PubMed, and Google Scholar using predefined terms; after screening and eligibility checks, 75 peer‑reviewed studies were included. Dominant techniques included Kernel Density Estimation (KDE), Getis–Ord Gi* clustering, crash rate analysis, and Empirical Bayes (EB) analyses, as well as machine-learning clustering. Across contexts, GIS consistently identified spatial blackspots, supported spatiotemporal trend analysis, and informed targeted countermeasures; key limitations were heterogeneous data quality, inconsistent methodological choices, and the integration of real‑time and behavioral data. GIS is effective for blackspot detection and decision support in road safety. Future work should prioritize standardizing methods, incorporating real‑time IoT streams and deep learning, and integrating behavioral and exposure data to improve prediction and intervention design.

The growing urban population and increasing climate anomalies pose persistent challenges to urban resilience by threatening food, water, and energy security and intensifying land-use competition. Utilizing urban rooftops for gardening, rainwater harvesting, and renewable energy systems offers a sustainable pathway to mitigate these pressures. This study develops a geospatial framework to evaluate the sustainable use of built-up areas in Coimbatore, India, using open-access geospatial datasets. Spatial and economic assessments were conducted to estimate the feasibility and revenue potential of rooftop gardening, rainwater harvesting, and photovoltaic installations. Our results reveal the economic potential of 368,748 rooftops in Coimbatore, with the projected revenue. Photovoltaic systems could generate ₹ 28.58 billion, while rooftop gardens and rainwater harvesting contribute ₹ 15.79 billion and ₹ 0.34 billion, respectively. Crop-specific analysis identified chillies as the most profitable rooftop crop, with a potential revenue of ₹ 38.51 billion, whereas coriander showed the lowest at ₹ 4.57 billion. These findings highlight the economic and environmental opportunities associated with rooftop agriculture and renewable energy systems, emphasizing their role in sustainable urban planning Open-access satellite imagery proved to be an invaluable tool in assessing the potential of rooftop spaces, offering valuable insights for urban planners and policymakers.

As cities strive for greater liveability, data-driven methods can enhance our understanding of public space behaviours and social interactions. Recent developments in computer vision technologies have significantly advanced the accuracy of micro-scale human behaviour detection, but there is a lack of methodologies that capture relational, nuanced behaviours within specific spatial and temporal environments. This paper presents the development of a computer vision and machine learning-based methodology to analyse co-presence and micro-social interactions in urban spaces, introducing new metrics for spatial behavioural analysis. The methodology was tested on a 22.5-min video dataset obtained at a university campus, demonstrating its capacity for trajectory analysis and detecting nuanced interpersonal behaviours including encountering, congregating, approaching and avoiding. Human observers validated the computer-generated behaviour classifications, achieving high agreement levels and demonstrating the system's accuracy in detecting diverse pedestrian interactions. The approach successfully offers fine-grained analysis of social behaviours and spatial patterns of co-presence, revealing how urban morphology influences social interaction hotspots. It advances environment-behaviour research by providing scalable, automated tools for detailed, data-driven analysis of public space vitality, with potential applications in urban design, social sciences, and policy-making.

The rapid expansion and urban development of cities have led to the widespread growth of highways and an increase in private vehicle usage. Consequently, traffic congestion and accidents have become significant concerns in cities like Tehran. To tackle these issues, it is crucial to identify traffic fluctuations(dynamicity points), which are critical for understanding urban transportation challenges. Traffic fluctuations represent sudden changes in traffic flow that signal potential road infrastructure problems, unexpected events, and traffic management inefficiencies. These dynamicity points, characterized by rapid transitions from light to heavy traffic, can reveal structural road design issues, accident-prone zones, and areas requiring targeted interventions. By utilizing location-based data collected from sensors and Google traffic maps, image processing techniques were employed to analyze traffic flow and identify areas with notable traffic fluctuations. A comparative analysis of these traffic fluctuations with existing accident data revealed a significant correlation between sections with high traffic fluctuations and driving accidents. Notably, approximately 70% of the accidents during the study period occurred within the vicinity of the identified dynamicity points. This study introduces a novel approach for calculating the geographical coordinates of high-potential traffic fluctuations, which can provide valuable insights for implementing targeted interventions to alleviate traffic congestion and enhance traffic safety.

Human behavior changes especially when facilitated and amplified by policy changes during situations such as the COVID pandemic can have far reaching consequences for urban mobility. We have seen orphaned roads and empty metro cars during times when “rush hour” should have been the norm. To better comprehend these phenomena, this work uses Agent-based modeling (ABM), specifically the MATSim framework, in combination with a wealth of publicly available data (CENSUS and Openstreetmap - OSM) to model pre- and during COVID urban mobility for the Washington, DC metro area. The available CENSUS data combined with population generation algorithms and MATSim allows us to model a population of four million people and their daily mobility patterns on a multimodal transportation network that includes a road network, the metro system, and a number of bus services in the Washington, DC metro area. In comparing the simulation output with ground-truth flows, we show that indeed our approach is capable to accurately capturing traffic flows in this multimodal network as observed before and during the COVID pandemic. Overall, this work demonstrates that publicly available population data (CENSUS) and transportation infrastructure and POI data (OSM) can be leveraged in a powerful simulation framework to accurately model urban mobility. Example scenarios could be the evaluation of future policy proposals as well as infrastructure projects that leverage mobility patterns.

The transformation of global urban areas has given rise to a strategic need for city branding, especially in satellite cities in developing countries that serve as extensions of major metropolitan cities. However, a significant gap remains in understanding how spatial centrality and public acceptability interact in shaping a city brand, especially in satellite cities at the transitional stages of development, as evidenced by Balikpapan, the satellite city of Indonesia's New Capital, Nusantara. This paper investigates the development of satellite city identity through a dual approach, focusing on spatial centrality and city brand acceptability. Drawing on Spatial Design Network Analysis (SDNA) metrics and perceptual indicators derived from Principal Component Analysis (PCA), the research investigates how geographic structure and emotional engagement collectively influence city identity. The five fundamental components of city brand acceptance consist of experiential and emotional attachment, awareness of city identity, infrastructure and comfort, urban environment and safety, and willingness to stay in the long term. However, a discrepancy exists between spatial aspects and city brand acceptance, as areas with high levels of accessibility and comfort do not necessarily correspond to a strong sense of city identity. Conversely, peripheral areas with lower spatial centrality may exhibit stronger emotional ties. This study recommends urban planning strategies that require central areas to be symbolically reinforced through narrative design and spatial formation, while peripheral areas with emotional resonance need to be preserved through improved connectivity to strengthen sustainable satellite city branding. This article contributes methodologically to the integration of spatial data computation and perception, with practical implications for adaptive city branding policies, particularly in competition with other satellite cities.

The use of telehealth has significantly increased in recent years, providing patients with remote access to healthcare services. To measure accessibility of telehealth services, two-step virtual catchment area (2SVCA) methods have been developed as a broadband-aware extension of the traditional two-step floating catchment area (2SFCA) framework. Although telehealth utilization varies across demographic groups, existing 2SVCA approaches do not account for such variation and typically treat population demand as uniform. To address the knowledge gap, this study proposes the demographic-weighted 2SVCA (DW-2SVCA), which enhances the 2SVCA framework by adjusting telehealth demand through a weight matrix based on demographic characteristics, including age, gender, race, and ethnicity. In a case study of primary healthcare in Mecklenburg County, North Carolina, we applied the proposed method to measure telehealth accessibility and compared it with the existing 2SVCA approach. While the overall statistical and spatial patterns were similar, DW-2SVCA exhibited greater variance in accessibility values. To further evaluate whether the proposed method more effectively captures demographic variations in telehealth utilization, we classified the study area into four groups based on demographic profiles by using hierarchical clustering and conducted statistical tests. The results revealed that the DW-2SVCA method indicates more pronounced and statistically significant differences, with higher accessibility in regions characterized by white and middle-aged populations and lower accessibility in more racially/ethnically diverse areas. This study offers a more comprehensive tool for evaluating telehealth accessibility, highlighting the importance of incorporating demographic variations in telehealth utilization.

Trajectory representation learning transforms raw trajectory data (sequences of spatiotemporal points) into low-dimensional representation vectors to improve downstream tasks such as trajectory similarity computation, prediction, and classification. Existing models primarily adopt self-supervised learning frameworks, often employing models like Recurrent Neural Networks (RNNs) as encoders to capture local dependency in trajectory sequences. However, individual mobility within urban areas exhibits regular and periodic patterns, suggesting the need for a more comprehensive representation from both local and global perspectives. To address this, we propose TrajRL-TFF, a trajectory representation learning method based on time-domain and frequency-domain feature fusion. First, considering the heterogeneous distribution of trajectory data in space, a quadtree is employed for spatial partitioning and coding. Then, each trajectory is converted into a quadtree-code based time series (i.e., time-domain signal), with its corresponding frequency-domain signal derived via Discrete Fourier Transform (DFT). Finally, a trajectory encoder, combining an RNN-based time-domain encoder and a Transformer-based frequency domain encoder, is constructed to capture the trajectory’s local and global features, respectively, and trained by a self-supervised sequence encoding-decoding framework with trajectory perturbation-reconstruction task. Experiments demonstrate that TrajRL-TFF outperforms baselines in downstream tasks including trajectory querying and prediction, confirming that integrating time- and frequency-domain signals enables a more comprehensive representation of human mobility regularities and patterns, which provides valuable guidance for trajectory representation learning and trajectory modeling in future studies.

With urban densification and the proliferation of high-rise structures, residents’ apartment views are getting obstructed from surrounding nature, especially greenery. Existing approaches often rely on simplified proxies or aggregated building- or floor-level metrics which does not capture individual-level variation. Most of them use coarse spatial data or subjective self-reports, lacking the granularity and precision to quantify the greenery visible to each resident from their own living space. This study introduces a conceptual and methodological framework for objectively modelling green views at the individual apartment level. Our Apartment Greenery View Measure was developed and assessed by (1) geocoding individual observer positions at window-level within apartment buildings, (2) implementing GIS-based three-dimensional viewshed analysis using high-resolution environmental datasets to objectively quantify views, and (3) examining agreement between modelled views and 445 residents’ self-reported perceptions using the green-to-grey ratio. The method was applied to 30 apartment buildings across Melbourne, Australia. Findings reveal variability in green view exposure by building height, floor-level, and apartment orientation. A moderate correlation (r = 0.556, ICC = 0.521) shows the agreement between objective and perceived view measures, with 39.1% participants overestimating and 60.9% underestimating their views. This underscores the need for objective, standardised measures that move beyond perception alone. The workflow supports aggregation at multiple spatial scales, from individual units to floors and buildings, providing a flexible framework for assessing visual green equity citywide. This provides a scalable, low-cost tool for planners, designers, and health researchers seeking to integrate visual greenery into urban housing, policy, and equity-focused interventions.

In February 2021, Winter Storm Uri severely impacted much of the southern United States, triggering unprecedented large-scale power outages. Recognizing that a similar extreme weather event could occur in the future, this study identifies as its primary research objective the development of a baseline power outage prediction model specifically tailored for the southern region of the United States. Central to this objective is the research question: Which variables and which regression models play the most significant role in accurately predicting power outages in this context? Given that large-scale outages are, in essence, a direct result of imbalances between electricity supply and demand, population was considered a key influencing factor. Furthermore, to ensure the model adequately reflects the meteorological characteristics of winter storms, several atmospheric variables—such as dew point and atmospheric pressure—were incorporated into the analysis. These variables are intended to capture the environmental dynamics that underpin outage occurrence during extreme cold events. Four machine learning models—Random Forest, eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost)—were employed in this study. In addition, to enable a comparison between these four machine learning approaches and traditional statistical models, Ridge regression and Lasso regression were also implemented, utilizing population and geographic information data in conjunction with meteorological variables to achieve this objective. To determine the optimal model configuration, Bayesian optimization was employed using tenfold cross-validation. The results revealed that XGBoost achieved the highest performance, with an R² score of 0.92. Furthermore, when the XGBoost model was utilized for prediction, a permutation importance analysis identified population, dew point, and pressure—in that order—as the most influential variables. Additionally, given that the number of data points varied by state during the test evaluation phase, a weighted evaluation metric was also computed using the data counts for each state. Under this weighted evaluation, XGBoost still achieved the highest R² score (0.74), further underscoring its robustness across heterogeneous state-level datasets. Consequently, this paper developed a foundational baseline model for power outage prediction due to winter storms in the Southern United States and identified essential variables through analysis.

Mobility cross spatial units represents the embodiment of how people manage activities between locations along temporal sequences. Spatiotemporal pattern nevertheless interacts with the socioeconomic characteristics of respected origin (push factors) and destination (pull factors) which widely discussed in spatial interaction literature. Observing this dynamics at higher spatial resolution allows us to entangle multifaceted nature of city, its complexity as a system or network, and the way it shapes movement of people. This study explore the extent interconnected elements of urban system or urban networks, in parallel with the appearance of external shock namely COVID outbreak, may affect estimation of mobility flows. To improve predictive power, Gravity Model is extended to Urban System Model by augmenting the complexities of urban network based on micro-analytical approach (intra-city networks). Our findings reveals better performance of a more complex Urban System Model as to compared with Gravity Model. Here, we leverage stratification in mobility by specifying mobility flows with respect to income status of respected areas. The occurrence of COVID outbreak followed by lockdown measure increases intra-class mobility, indicating the coupling between socioeconomic distance and geographical distance. Flows between areas with similar economic ranges are more predictable than the one of different level. Furthermore, the presence of pull factors is more affluent than push factors in determining mobility regardless the severity of external shock.

Accurate classification of urban sprawl is vital for sustainable urban planning, yet most regional-scale approaches overlook local spatial heterogeneity and lack robust validation. This study presents a comprehensive framework that integrates high-resolution sliding-window analysis, advanced spatial metrics, Morphological Spatial Pattern Analysis (MSPA) and building density for validation, and machine learning-based feature importance assessment. The framework is applied to both developing cities (Colombo and Kandy, Sri Lanka) and a developed city (Hong Kong) for the years 2005, 2015, and 2025. Twenty spatial metrics are computed within 510 m × 510 m windows, with the optimal window size determined through sensitivity analysis, and Pearson correlation used for dimensionality reduction. Urban sprawl typologies are extracted via K-means clustering, with the optimal cluster number determined by the Gap Statistic and clustering quality evaluated using Silhouette scores. Metric weighting is performed using CRITIC (Criteria Importance Through Intercriteria Correlation), which prioritizes metrics based on their discriminative power and independence. Five distinct sprawl types: infill, extension, linear, clustered, and leapfrog, are identified and validated against MSPA-derived morphological elements and building density. Random Forest and Cliff’s Δ analyses highlight transport infrastructure, especially road density and proximity to main roads, as the primary drivers of sprawl, alongside population density and topography. The framework demonstrates robust predictive performance and offers a scalable, locally adaptive tool for precise urban sprawl classification, supporting evidence-based planning and policy.

Rapid urbanization, increased motorization, and industrialization have led to ever-increasing levels of Traffic-Related Air Pollution (TRAP), which has significant implications for public health and urban sustainability. This systematic review assesses the application of Geographic Information Systems (GIS) to model vehicle emissions and the related health impacts in urban areas. This review is based on literature published between 1990 and 2024. We screened 4,780 peer-reviewed articles and 780 met inclusion criteria. We examined the computational methods used in impact studies, including data from spatial datasets, pollutant variables, and epidemiological data. The most common methods were geo-statistical interpolation (Kriging, Geographically Weighted Regression), Land-Use Regression (LUR), and machine learning (Support Vector Regression, Neural Networks), typically with California Line Source Dispersion Model (CALINE) and Community Multiscale Air Quality Model (CMAQ). To pull multiple analytical perspectives, we purposefully combined systematic review methods with techniques of bibliometric analysis using VOS-viewer and R-software, allowing us to the research output and trends, collaborative networks and research themes. Ultimately our mixed-methods approach demonstrated important differences between developed and developing contexts regarding data availability, exposure modeling approaches and the integration of health co-benefits from active transport. Building on these findings, we introduce a GIS-based decision-support framework integrating traffic data, remote sensing, pollution modeling and health monitoring into a real-time, open-access platform to assist with evidence-based urban planning. This review, emphasizing the computational tools to create high-resolution exposure maps and better translate policy into practice, advances the field of computational urban science and provides a reproducible framework for ameliorating pollution-related health impacts in their best-case scenario rapidly urbanizing cities.

This study introduces an integrative approach to forecasting urban morphogenesis and housing allocation dynamics in rapidly developing cities. Doha metropolitan is examined as the case study, with projections extending to the year 2035. Based on a robust theoretical framework, the study considers three factors: (i) residential land capacity and regulatory policies, (ii) housing supply and demand, and (iii) housing preferences. An integrative methodological framework was adopted, including content analysis of housing regulatory frameworks, a survey of housing preferences, and Machine Learning (ML). A Convolutional Long Short-Term Memory (ConvLSTM) network was utilized to simulate parametric urban morphogenesis. The integrated model was then employed to project housing allocation patterns for the year 2035. Spatial stratification in ArcGIS-Pro revises the predicted parametric urban morphogenesis, focusing on the residential growth within regulated growth boundary. The findings show that housing demand and supply patterns in Doha metropolitan are shaped by regulatory policies and socioeconomic factors. Predicted growth is concentrated in suburban areas, dominated by low-density villas, while the waterfront will require inclusionary zoning to balance housing needs. The study offers policymakers, urban planners, and developers a robust tool for regulatory adaptation that aligns urban morphology with housing market dynamics. By providing a nuanced prediction, the model supports sustainable urban growth by balancing housing needs for a diverse population and addressing supply and demand imbalances. This study provides a valuable reference for cities facing similar growth challenges worldwide.

Using Jiangsu Province as a case study, this paper analyzed the spatial distribution characteristics of intangible cultural heritage (ICH) in Jiangsu, China. It employed the nearest neighbor index (NNI), geographic detectors, and related analysis methods to investigate the association between ICH and local tourism. It was found that ICHs in Jiangsu was mainly concentrated in the southern part of the province, particularly in Nanjing, Suzhou, and Yangzhou. In terms of category distribution, the ICH of Jiangsu mainly comprises traditional skills, 38 of which are national-level ICHs. Moreover, the overall concentration degree of ICH in Jiangsu was relatively high (NNI = 0.268), and traditional skills exhibited the highest density (NNI = 0.362). Geographic detector results indicated that the spatial distribution of ICH in Jiangsu was most strongly influenced by gross domestic product (GDP) (q = 0.823), followed by the number of historical and cultural cities (q = 0.764). Correlation analysis showed that ICH was significantly associated with various tourism indicators. These results highlight the spatial distribution characteristics of Jiangsu’s ICH and indicate that its international development remains insufficient.

Existing indoor spatial keyword path queries have not yet addressed user exclusion preferences, and research that simultaneously considers time constraints and exclusion preferences remains unexplored. To address this issue, this paper proposes an indoor spatial keyword path query method based on time awareness and exclusion preference (TEISKPQ). The method introduces a novel and purpose-built Time-awareness and Exclusion-preference Indoor Spatial Keyword tree (TEISK-tree) index structure, specifically designed to organize and manage indoor spatial keyword objects and their associated information. Based on the TEISK-tree index, an index-driven pruning algorithm is developed to rapidly eliminate nodes that do not meet user requirements, significantly reducing computational complexity and enhancing query efficiency. Furthermore, a greedy-based initial path generation algorithm is proposed, which employs a weighted evaluation function to comprehensively assess spatial, textual, and temporal factors between nodes, thereby generating multiple locally optimal initial paths. Finally, an improved genetic algorithm is designed to perform global search-based optimization of the initial paths, thereby better satisfying user requirements by helping to escape local optima and increasing the likelihood of finding high-quality solutions. Experimental results demonstrate that the proposed method exhibits high efficiency and good convergence speed in complex indoor environments, which can improve the efficiency and practicality of indoor path query in urban smart environments and provide support for efficient and intelligent spatial information retrieval.

Prosperous waterway economics require rigorous safety measures. Unmanned aerial vehicle (UAV) offers massive images of inland waterways, within which navigation mark detection plays a critical role in ensuring waterway safety. This paper proposes a deep learning-based method for detecting navigation marks in UAV images. Firstly, a dataset of inland waterway navigation marks is constructed from UAV aerial images, which includes data collection, image enhancement, sample creation, and sample annotation. Secondly, a deep learning network model is developed, which uses ResNet-50 as the backbone, incorporates Coordinate Attention and Large-Scale Selective Kernel Attention mechanisms, integrates a Feature Pyramid Network (FPN) for feature enhancement, and uses Distance Intersection over Union (DIoU) as the loss function. Thirdly, the model is trained and evaluated on the constructed dataset, followed by precision assessment and post-processing. This paper explore a deep learning network model for small object detection in UAV images and establish a comprehensive workflow for detecting inland waterway navigation marks, thereby providing technical support for waterway safety.

We introduce the first natural language interface for complex urban analytics, leveraging Large Language Models (LLMs) and Spatio-Temporal Transactional Networks (STTNs). By combining intuitive natural language querying with structured data analytics, our framework simplifies complex urban analyses, such as identifying commuter patterns, detecting anomalies, and exploring mobility networks. We propose a comprehensive evaluation dataset that demonstrates that minor architectural improvements can significantly improve analytical accuracy. Our approach bridges the gap between non-expert users and sophisticated urban insights, paving the way for accessible, reliable, and scalable urban data analytics.

Crime prevention requires accurate prediction of the spatial and temporal distribution of criminal activities to effectively allocate law enforcement resources. However, many trending crime prediction algorithms lack comprehensive spatio-temporal structures and often consider only single input variables. This study innovatively using in ST-Cokriging method integrated both historical crime records as the primary variable and crime-related geo-tagged Twitter data as the co-variable for crime prediction. The predictive method has been specifically developed to assess crime risk across three major crime types—street crime, property crime, and vehicle crime—and applied in the San Francisco Bay Area (SFBA), California, a region characterized by high development and heightened crime sensitivity, for both prediction and validation. The results indicate that incorporating social media data into a spatio-temporal statistical method improves the associations between predicted and actual crime risk, reduced the Root Mean Squared Error (RMSE), and enhanced the identification of crime risk areas for both weekdays and weekends across three crime types compared to the method without the co-variable. This study presents a new multi-variable approach to more accurately predict crime, enabling law enforcement proactively address crime of varying nature in urban areas.

Intra-urban migration plays a crucial role in shaping urban structure and socio-economic dynamics. Most existing studies rely on small-scale survey data or have a coarse spatial resolution, making it difficult to conduct detailed network analysis at the urban scale to fully understand the complexity and dynamics of migration patterns. To address the gap, this study conducted a subdistrict-level fine-grained network analysis, involving more than 800,000 relocation data with detailed demographic and housing information at subdistrict levels in Shenzhen in 2015, to explore the overall relocation patterns and the relocation differences among different groups. The findings reveal that short-distance relocations dominate, with major hubs serving as central points of population flow in the study area (e.g., Gongming and Shajing areas). The relocation patterns also indicate specific pathways guiding movement between city areas. Moreover, demographic factors such as marital status, education level, and age significantly influence relocation behaviour. For instance, elderly individuals move infrequently, but when they do, they often relocate over longer distances. Men tend to migrate to diverse areas, while women prefer similar ones. Highly educated individuals move longer distances, typically within economic core areas. Overall, our study provides new perspectives for understanding the complex mechanisms of intra-urban population migration.

Criminal trajectory reconstruction is a crucial area of study in the investigative and evidentiary processes of public security departments. This paper proposes a method for reconstructing criminal trajectories based on a mobile reference system, starting from sparse location data obtained from surveillance. The goal is to dynamically balance safety and danger perception anchor points, integrating travel distance to reconstruct trajectories that align with the anti-surveillance behavior of criminals. First, the characteristics of criminal movement are analyzed to construct a criminal mobile reference system, where the reference range constrains the potential scope of missing trajectories. Spatial elements influencing the trajectory are selected as anchor points. Based on this reference system, an improved heuristic algorithm balances anchor point transition probabilities and travel distance to identify the optimal sequence of anchor points, filling in the gaps between sparse location points. Experimental comparisons demonstrate that the proposed method reconstructs more accurate and reasonable criminal trajectories. The research provides theoretical support for public security investigations, analyzing criminal movement characteristics and using spatial points related to criminal travel to supplement missing trajectories, addressing gaps in criminal trajectory reconstruction research and contributing to the accurate reconstruction of criminal movement paths.

Internet of Things (IoT) communication technology is widely used in industry. IoT LPWAN technology is used to build a binocular vision 3D reconstruction system for garden scenes to improve the problem of insufficient 3D image construction of garden scenes. By analyzing the imaging principle of binocular vision, the camera calibration method is optimized, and the binocular vision model is constructed. The feature processing and extraction of binocular vision are key to 3D scene construction, but traditional binocular vision systems have always faced difficulties in scene feature extraction, which affects the composition effect of the scene. Therefore, on the basis of the traditional SURF feature extraction algorithm, a SURF-B matching algorithm combining LDB feature description is proposed for extraction of image feature information. This performance experiment showed that in multi-view image feature matching, the feature matching errors of SIFT, SURF, and the proposed SURF-B algorithms were 125, 100, and 45, and the matching errors were 0.220, 0.115, and 0.036, respectively. At the same time, in the multi-algorithm matching accuracy test, the proposed SURF-B algorithm also had excellent matching accuracy and convergence performance. The research content has important reference significance for improving the composition effect of the garden scene and the layout effect of the garden landscape.

This work proposes an opinion dynamics model describing public interactions on a given issue of public interest, with opinion leaders expressing changing support or opposition over time. Motivated by a system of ordinary differential equations from prior work, extensions were introduced accounting for the degree and direction of opinion leaders’ support, including the time-dependent parameters associated with their capacities to affect public opinion. Aside from these advances, the proposed model defines the degree of support of opinion leaders as a multi-criteria concept, a more realistic and comprehensive representation of their influence. The proposed dynamical system was applied in a case study modelling public opinion on a bus rapid transit (BRT) project. The model parameters linked to the interactions of sub-populations were adopted from a previous study. Meanwhile, archival data were extracted to proxy the influence capacities of opinion leaders and their degree of support under a specific criterion. Operations of intuitionistic fuzzy sets, more generalized sets that handle data ambiguity, were implemented to generate multi-criteria support (or opposition) degrees of opinion leaders over time. Findings suggest the following: (1) in the absence of opinion leaders, the public becomes indifferent about their opinion on the BRT project, (2) public opinion tends to be highly influenced by opinion leaders, and (3) intervention of opinion leaders results in a “polarizing effect”, where neutral sub-population dissipates in favor of the agree or disagree sub-population. These findings help determine the level of public support for a given project in the presence of opinion leaders.

Amidst rapid advancements in artificial intelligence and smart city technologies, this paper argues that the Metaverse, as a virtual form of smart cities, offers the potential to advance Sustainable Development Goals (SDGs), particularly in education, innovation, and sustainable urban development. However, goals related to inequality reduction and climate action are underrepresented. We argue that the Metaverse holds transformative potential to advance the SDGs, but its trade-offs for equality and environmental sustainability must be carefully considered. This requires adopting ethical and inclusive governance frameworks that are based on systems thinking.

This study aims to assess the microclimatic differences between multiple urban areas in a university campus and to examine the impact of adding trees to these different areas at the University of Sharjah's campus in the United Arab Emirates (UAE). Five locations were chosen and compared to each other by first collecting weather data for 24 h on November 25–26, 2024, a month characterized by a rise in student activities and a transition between the hotter and the cooler seasons in the UAE. These measurements are used to calibrate subsequent Envi-met simulations. Tree types that are native or adapted to UAE’s hot climate are added to all sites under study, and the microclimatic conditions before and after their addition were compared for each site individually and collectively. According to the study's results, variations exist across the five studied campus sites, with open spaces experiencing higher thermal stress due to direct solar exposure and insufficient shading. Courtyards, or areas that resemble courtyards, exhibit better thermal conditions due to self-shading effects. While air temperature across the sites fell by less than 1 °C, adding trees that offered shade led to MRT decreases of up to 41%. These findings emphasize the role of vegetation in improving the outdoor microclimate by enhancing shading.

Understanding the relationship between urban growth and CO2 emissions is essential for sustainable urban and environmental planning in China. Even though some studies have been conducted in this regard, there is a lack of comprehensive studies that integrate socioeconomic and nighttime light (NL) data on both spatial and temporal scales. Therefore, using NL data as a proxy for urban growth, this study offers a novel approach to assess city size distribution (CD) and CO₂ emission dynamics from 2000 to 2020 at the provincial and prefecture levels. The present study was conducted in three phases: (1) assessing the association between urban growth and socioeconomic characteristics; (2) measuring CD dynamics using corrected NL data; and (3) modeling CO₂ emission dynamics through panel data analysis. While the Ordinary Least Squares (OLS) method examined the relationship between socioeconomic characteristics and urban growth, the CD dynamics were measured using Catteow’s formula. A panel unit root test, panel co-integration test, and panel regression analyses were performed to explore the relationship between urban growth and CO₂ emissions. Results revealed that maximum NL data have stronger correlations with population, GDP, and EPC at the provincial level than at the prefecture level, with an average R2 range from 0.6219 to 0.8985. The analysis of CD dynamics revealed an increase in urban disparity, particularly among larger cities, with the q value rising from 0.7920 to 0.8268. CO₂ emissions expanded by 250.76% from 2000 to 2020, with the highest growth seen in coastal megacities. Panel unit root and co-integration tests confirmed a long-term relationship between urban growth and CO₂ emissions at both scales. Panel regression analysis showed a positive and significant impact of urban growth on CO₂ emissions at the national level and across all regions and provinces. These findings highlight the importance of sustainable urban planning strategies that incorporate socioeconomic characteristics with spatial and temporal considerations to reduce CO₂ emissions in China. However, further research is necessary to explore multidimensional strategies for balancing urban expansion and CO₂ emissions.

The study investigates the impact of the land use changes on the urban heat island effect ratio (UHIER), focusing on the urban development fringe of Ankara, Türkiye. Initially characterized by rural land uses the areas has experienced significant transformations into residential estates, mostly including high-rise blocks and low-rise villas. Urban development patterns in 2013 and 2023 were compared with changes in UHIER and local climate zone classes (LCZCs) using RS and GIS techniques for UHIER calculation, and the World Urban Database and Access Portal Tools (WUDAPT) protocol for LCZ mapping. Overall, UHIER values have a tendency to rise, as areas with increaing UHIER are found to be twice as large as those with decreasing UHIER. Increasing UHIER is highly associated with increases in open high-rise and sparsely built areas, accompanied by decreases in low plants. UHIER, on the other hand, is mosly characterized by a reduction in large low-rise built-types. The parts where UHIER remains unchanged suggests that although compact high-rise, open high-rise, and sparsely built areas have increased, the reduction in other built types—particularly large low-rise areas—along with a rise in tree density, appears to balance these changes. Therefore, to prevent high UHI impact when the area is fully developed, more landscaping features, particularly trees, can be integrated and mid-rise and low-rise developments can be preferred over high-rises, ensuring the efficient land use.

This paper introduces two novel deep generative frameworks for synthetic population generation that jointly model household and individual attributes. In leveraging Variational Autoencoders (VAEs), we propose herein the SVAE-Pop2 method, which employs a single VAE with fixed-size padded inputs, along with the MVAE-Pop2 method, which uses dedicated models for various household sizes. Evaluated on a French household travel survey dataset, our experiments reveal that while both approaches effectively reproduce the actual population’s characteristics, MVAE-Pop2 achieves greater fidelity in joint attribute distributions. The proposed methodologies suggest improvements in agent-based simulations and urban modeling by means of generating realistic, multi-layered synthetic populations.

Accurate delineation of urban spatial extent is essential for planning, yet conventional methods based on land cover and satellite imagery are often time-consuming and may lag behind urban changes. This study explores how urban functional area can be delineated using Points of Interest (POI) data and Kernel Density Estimation (KDE), offering an activity-based alternative to morphology-based approaches. Using Pekanbaru as the study area, a metropolitan city in Indonesia, the method incorporates spatial autocorrelation to weight POIs and generate a KDE surface. The resulting delineation is compared to Sentinel-2-derived built area using the STEP Similarity Index and Jaccard Index. STEP results indicate strong thematic (0.96) and positional (0.97) similarity, with low shape and edge values, showing that POI-based KDE captures activity intensity rather than physical form. The Jaccard Index (0.64) confirms a moderate spatial overlap. While satellite data reflects built structures, KDE highlights zones of concentrated human activity, supporting its utility for planning applications. Future work should advance POI temporal filtering, KDE threshold calibration, and functional zone mapping, enabling integration into multi-scale spatial planning. This study contributes a scalable, data-driven method for delineating urban extent using openly available activity-based data.

The architect Christopher Alexander made major contributions in architecture, city science, computer science and other fields. His earlier works, notably “A City is Not a Tree,” contributed to the understanding of cities as complex adaptive systems. Less well known, although no less theoretically ambitious, was Alexander’s later work after the year 2000. Here we explore the potential contribution of this work to contemporary urban challenges, focusing on Alexander’s magnum opus The Nature of Order: An Essay of the Art of Building and the Nature of the Universe. We assess its theoretical contribution to the field of sustainable urbanism, with a focus on its computational aspects, as well as the critique it poses for conventional approaches to urban sustainability.