Computers Environment and Urban Systems最新文献_第4页

Identifying street multi-activity potential (SMAP) and local networks with MLLMs and multi-view graph clustering 利用mllm和多视图图聚类识别街道多活动潜力（SMAP）和本地网络

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-09-02 DOI: 10.1016/j.compenvurbsys.2025.102350

Jiatong Li , Mingyi Ma , Yuan Lai

Streets are essential public spaces hosting a variety of social, cultural, and economic activities that collectively form urban vitality. However, due to limitations in research methodology and data, existing studies often oversimplify street activities by focusing solely on pedestrian flows. This study introduces a novel approach using Multimodal Large Language Models (MLLMs) and multi-view graph-based community detection to systematically evaluate street multi-activity potential (SMAP). Utilizing diverse urban data, we quantified the SMAP based on six common pedestrian activities (sitting, standing, walking, jogging, exercising, and street vending) in Beijing's central urban area. Results reveal significant spatial disparities in the suitability scores of different activity types, challenging the conventional reliance on walking as a proxy for street activities. By applying community detection algorithm with multi-view graph fusion and reinforcement learning, we identified 245 SMAP areas and uncovered their underlying spatial network patterns in Beijing. Assessment of SMAP areas' total potential and diversity of potential reveals the complex relationship between the two dimensions. By further identifying high total potential SMAP areas with varied levels of diversity, we discovered their distinct patterns in semantic features and spatial distributions. Overall, this study develops a novel and scalable framework for evaluating street spaces and observing their potential for diverse activities, which will guide future planning to support activity diversity and enhance urban vitality.

街道是举办各种社会、文化和经济活动的重要公共空间，共同形成城市活力。然而，由于研究方法和数据的限制，现有的研究往往过于简化街道活动，只关注行人流量。本文介绍了一种利用多模态大语言模型（MLLMs）和基于多视图图的社区检测来系统评估街道多活动潜力（SMAP）的新方法。利用不同的城市数据，我们基于北京中心城区六种常见的步行活动（坐、站、走、慢跑、锻炼和街头贩卖）对SMAP进行了量化。结果显示，不同活动类型的适宜性得分存在显著的空间差异，挑战了传统的以步行为代表的街头活动。采用基于多视图图融合和强化学习的社区检测算法，对北京市245个SMAP区域进行了识别，揭示了其潜在的空间网络格局。SMAP地区的总潜力和潜力多样性评价揭示了两者之间的复杂关系。通过进一步识别具有不同多样性水平的高总潜力SMAP区域，我们发现了它们在语义特征和空间分布上的独特模式。总体而言，本研究开发了一个新颖的可扩展框架，用于评估街道空间并观察其多样化活动的潜力，这将指导未来的规划，以支持活动多样性并增强城市活力。

{"title":"Identifying street multi-activity potential (SMAP) and local networks with MLLMs and multi-view graph clustering","authors":"Jiatong Li , Mingyi Ma , Yuan Lai","doi":"10.1016/j.compenvurbsys.2025.102350","DOIUrl":"10.1016/j.compenvurbsys.2025.102350","url":null,"abstract":"<div><div>Streets are essential public spaces hosting a variety of social, cultural, and economic activities that collectively form urban vitality. However, due to limitations in research methodology and data, existing studies often oversimplify street activities by focusing solely on pedestrian flows. This study introduces a novel approach using Multimodal Large Language Models (MLLMs) and multi-view graph-based community detection to systematically evaluate street multi-activity potential (SMAP). Utilizing diverse urban data, we quantified the SMAP based on six common pedestrian activities (sitting, standing, walking, jogging, exercising, and street vending) in Beijing's central urban area. Results reveal significant spatial disparities in the suitability scores of different activity types, challenging the conventional reliance on walking as a proxy for street activities. By applying community detection algorithm with multi-view graph fusion and reinforcement learning, we identified 245 SMAP areas and uncovered their underlying spatial network patterns in Beijing. Assessment of SMAP areas' total potential and diversity of potential reveals the complex relationship between the two dimensions. By further identifying high total potential SMAP areas with varied levels of diversity, we discovered their distinct patterns in semantic features and spatial distributions. Overall, this study develops a novel and scalable framework for evaluating street spaces and observing their potential for diverse activities, which will guide future planning to support activity diversity and enhance urban vitality.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"122 ","pages":"Article 102350"},"PeriodicalIF":8.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144932759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hyperlocal heat stress around bus stops in Philadelphia: Insights from spatio-temporal microclimate modeling and explainable AI 费城公交车站周围的超局部热应力：来自时空微气候模型和可解释人工智能的见解

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-09-02 DOI: 10.1016/j.compenvurbsys.2025.102341

Shengao Yi , Xiaojiang Li , Donghang Li , Xinyu Dong , Ruoyu Wang , Qian Xu

The Urban Heat Island (UHI) effect significantly impacts public transit users, particularly those waiting at bus stops, where prolonged exposure to extreme heat poses health risks. Despite increasing attention to climate resilience, limited research has focused on hyperlocal, pedestrian-level thermal stress at bus stops or its relationship with the surrounding urban environment. To address this gap, we generated hourly 1-meter resolution Universal Thermal Climate Index (UTCI) maps for Philadelphia using high-resolution, multi-source geospatial data and microclimate modeling, capturing detailed summer daytime spatio-temporal heat stress patterns around more than 8,000 bus stops. We further developed an explainable machine learning framework, combining Random Forest (RF) and SHAP analysis to uncover complex, nonlinear relationships and threshold effects between heat stress and both built environment and socioeconomic variables. Key findings include: (1) Significant spatio-temporal variation in heat stress, with consistently high levels at midday across the city; (2) Higher heat stress around bus stops located in low-income neighborhoods, while more affluent areas (e.g., higher median household value) exhibit reduced thermal exposure; (3) Green View Index (GVI) and Enclosure emerged as the most effective heat-mitigating features, and (4) complex threshold effects across key urban indicators highlight the importance of targeted and equitable interventions to reduce heat stress in vulnerable areas.

城市热岛效应严重影响公共交通用户，特别是那些在公交车站等待的人，他们长期暴露在极端高温下会带来健康风险。尽管人们越来越关注气候适应能力，但有限的研究集中在公交车站的超局部、行人水平的热应力或其与周围城市环境的关系上。为了解决这一差距，我们利用高分辨率、多源地理空间数据和小气候建模，为费城生成了每小时1米分辨率的通用热气候指数（UTCI）地图，详细捕捉了8000多个公交车站附近夏季白天的时空热应力模式。我们进一步开发了一个可解释的机器学习框架，结合随机森林（RF）和SHAP分析来揭示热应力与建筑环境和社会经济变量之间复杂的非线性关系和阈值效应。主要发现包括：(1)热应激的时空差异显著，中午时全市热应激水平持续较高；(2)低收入社区公交车站周围的热应力较高，而较富裕地区（如家庭中位数较高）的热暴露程度较低；(3)绿色景观指数（GVI）和围护是最有效的热缓解特征；(4)城市主要指标的复杂阈值效应突出了有针对性和公平的干预措施对减少脆弱地区热应激的重要性。

{"title":"Hyperlocal heat stress around bus stops in Philadelphia: Insights from spatio-temporal microclimate modeling and explainable AI","authors":"Shengao Yi , Xiaojiang Li , Donghang Li , Xinyu Dong , Ruoyu Wang , Qian Xu","doi":"10.1016/j.compenvurbsys.2025.102341","DOIUrl":"10.1016/j.compenvurbsys.2025.102341","url":null,"abstract":"<div><div>The Urban Heat Island (UHI) effect significantly impacts public transit users, particularly those waiting at bus stops, where prolonged exposure to extreme heat poses health risks. Despite increasing attention to climate resilience, limited research has focused on hyperlocal, pedestrian-level thermal stress at bus stops or its relationship with the surrounding urban environment. To address this gap, we generated hourly 1-meter resolution Universal Thermal Climate Index (UTCI) maps for Philadelphia using high-resolution, multi-source geospatial data and microclimate modeling, capturing detailed summer daytime spatio-temporal heat stress patterns around more than 8,000 bus stops. We further developed an explainable machine learning framework, combining Random Forest (RF) and SHAP analysis to uncover complex, nonlinear relationships and threshold effects between heat stress and both built environment and socioeconomic variables. Key findings include: (1) Significant spatio-temporal variation in heat stress, with consistently high levels at midday across the city; (2) Higher heat stress around bus stops located in low-income neighborhoods, while more affluent areas (e.g., higher median household value) exhibit reduced thermal exposure; (3) Green View Index (GVI) and Enclosure emerged as the most effective heat-mitigating features, and (4) complex threshold effects across key urban indicators highlight the importance of targeted and equitable interventions to reduce heat stress in vulnerable areas.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"122 ","pages":"Article 102341"},"PeriodicalIF":8.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144932757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reassessing air pollution exposure: How daily mobility and activities shape individual risk in greater Paris 重新评估空气污染暴露：日常流动性和活动如何影响大巴黎地区的个人风险

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-26 DOI: 10.1016/j.compenvurbsys.2025.102340

Taos Benoussaïd , Isabelle Coll , Hélène Charreire , Inès Makni , Malo Costes , Arthur Elessa Etuman

Understanding individual exposure to air pollution is essential for tackling environmental inequalities and informing public policies aimed at reducing disparities. Traditional approaches often focus on residential locations, but exposure is intrinsically linked to daily mobility, activities and socio-economic profiles. This study presents new results based on a dynamic exposure modelling approach that takes these dimensions into account, offering a more realistic assessment of air pollution risk. By integrating high-resolution air quality data with detailed information on individual mobility, activities and socio-economic characteristics, we quantify the exposure of 400,000 individuals in the Île-de-France region. Our approach takes into account all the environments that individuals visit during the day and the time spent in each of them, going beyond static exposure assessments based on residential location. We compare this dynamic model with traditional exposure calculations, revealing significant differences in the spatial distributions of PM10 and NO2 exposure. Our analysis highlights how mobility patterns and daily activities contribute to total exposure, demonstrating that place of residence is only one part of reality. For example, commuting, workplaces and leisure activities play a key role in determining individual exposure levels. The results of our study show that dynamic exposure calculation provides a better understanding of exposure factors and offers a framework for understanding environmental inequalities. By shifting the focus from home-based to person-based exposure, our approach makes it possible to identify levers for action to reduce disparities and support targeted public health action. Our study redefines the way in which we assess the risks associated with air pollution, by highlighting the need to take into account mobility behaviour and individual trajectories.

了解个人暴露于空气污染的情况对于解决环境不平等问题和为旨在减少不平等的公共政策提供信息至关重要。传统方法通常侧重于住宅地点，但暴露与日常流动性、活动和社会经济状况有着内在联系。这项研究提出了基于动态暴露建模方法的新结果，该方法考虑了这些因素，提供了更现实的空气污染风险评估。通过将高分辨率空气质量数据与个人流动性、活动和社会经济特征的详细信息相结合，我们量化了Île-de-France地区40万人的暴露情况。我们的方法考虑了个人在白天访问的所有环境以及在每个环境中花费的时间，超越了基于居住地点的静态暴露评估。我们将该动态模型与传统的暴露计算方法进行了比较，发现PM10和NO2暴露的空间分布存在显著差异。我们的分析强调了移动模式和日常活动对总暴露的影响，表明居住地只是现实的一部分。例如，通勤、工作场所和休闲活动在决定个人暴露水平方面起着关键作用。我们的研究结果表明，动态暴露计算可以更好地理解暴露因素，并为理解环境不平等提供一个框架。通过将重点从以家庭为基础的接触转移到以个人为基础的接触，我们的方法可以确定采取行动的杠杆，以缩小差距并支持有针对性的公共卫生行动。我们的研究通过强调考虑移动行为和个人轨迹的必要性，重新定义了我们评估空气污染相关风险的方式。

{"title":"Reassessing air pollution exposure: How daily mobility and activities shape individual risk in greater Paris","authors":"Taos Benoussaïd , Isabelle Coll , Hélène Charreire , Inès Makni , Malo Costes , Arthur Elessa Etuman","doi":"10.1016/j.compenvurbsys.2025.102340","DOIUrl":"10.1016/j.compenvurbsys.2025.102340","url":null,"abstract":"<div><div>Understanding individual exposure to air pollution is essential for tackling environmental inequalities and informing public policies aimed at reducing disparities. Traditional approaches often focus on residential locations, but exposure is intrinsically linked to daily mobility, activities and socio-economic profiles. This study presents new results based on a dynamic exposure modelling approach that takes these dimensions into account, offering a more realistic assessment of air pollution risk. By integrating high-resolution air quality data with detailed information on individual mobility, activities and socio-economic characteristics, we quantify the exposure of 400,000 individuals in the Île-de-France region. Our approach takes into account all the environments that individuals visit during the day and the time spent in each of them, going beyond static exposure assessments based on residential location. We compare this dynamic model with traditional exposure calculations, revealing significant differences in the spatial distributions of PM10 and NO2 exposure. Our analysis highlights how mobility patterns and daily activities contribute to total exposure, demonstrating that place of residence is only one part of reality. For example, commuting, workplaces and leisure activities play a key role in determining individual exposure levels. The results of our study show that dynamic exposure calculation provides a better understanding of exposure factors and offers a framework for understanding environmental inequalities. By shifting the focus from home-based to person-based exposure, our approach makes it possible to identify levers for action to reduce disparities and support targeted public health action. Our study redefines the way in which we assess the risks associated with air pollution, by highlighting the need to take into account mobility behaviour and individual trajectories.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"122 ","pages":"Article 102340"},"PeriodicalIF":8.3,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144894924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Toward satisfactory public accessibility: A crowdsourcing approach through online reviews to inclusive urban design 实现令人满意的公共可达性：通过在线评论的众包方法来实现包容性城市设计

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-23 DOI: 10.1016/j.compenvurbsys.2025.102329

Lingyao Li , Songhua Hu , Yinpei Dai , Min Deng , Parisa Momeni , Gabriel Laverghetta , Lizhou Fan , Zihui Ma , Xi Wang , Siyuan Ma , Jay Ligatti , Libby Hemphill

As urban populations grow, the need for accessible urban design has become urgent. Traditional methods for assessing public perceptions of accessibility, such as surveys and interviews, are often resource-intensive and geographically limited in scope. Crowdsourcing via online reviews offers a valuable alternative to understanding public perceptions, and advancements in large language models (LLMs) can facilitate their use. In this study, we examine over one million Google Maps reviews from points of interests (POIs) across the United States and fine-tune the Llama 3 model using the Low-Rank Adaptation (LoRA) technique to identify public sentiment toward accessibility. At the POI level, most categories – restaurants, retail, hotels, and healthcare – show negative sentiments, indicating persistent barriers across key sectors. Socio-spatial regression analysis reveals that more positive sentiment is associated with areas that have higher proportions of white residents and greater socioeconomic advantage. Conversely, more negative sentiment is expressed in areas with higher concentrations of elderly and highly-educated populations. Interestingly, no clear link is found between the presence of disabilities and public sentiments, but a significant positive relationship does exist between disability-friendly scores and public perception. Overall, our findings demonstrate the value of crowdsourcing with LLM-enhanced analysis in identifying accessibility challenges and informing inclusive urban design, offering actionable insights for planners, policymakers, and advocates striving toward more equitable cities.

随着城市人口的增长，对无障碍城市设计的需求变得迫切。评估公众对无障碍的看法的传统方法，如调查和访谈，往往是资源密集和地理范围有限的。通过在线评论的众包为理解公众的看法提供了一个有价值的选择，大型语言模型（llm）的进步可以促进它们的使用。在这项研究中，我们检查了来自美国各地兴趣点（poi）的100多万条谷歌地图评论，并使用低等级适应（LoRA）技术对Llama 3模型进行微调，以确定公众对可访问性的看法。在POI级别，大多数类别（餐馆、零售、酒店和医疗保健）表现出负面情绪，表明关键行业之间存在持续障碍。社会空间回归分析显示，白人居民比例高、社会经济优势大的地区，其积极情绪越高。相反，在老年人和高学历人口更集中的地区，负面情绪表达得更多。有趣的是，残疾的存在与公众情绪之间没有明显的联系，但残疾友好得分与公众感知之间确实存在显著的正相关关系。总体而言，我们的研究结果证明了通过法学硕士增强分析的众包在识别无障碍挑战和为包容性城市设计提供信息方面的价值，并为规划者、政策制定者和倡导者提供可操作的见解，以努力实现更公平的城市。

{"title":"Toward satisfactory public accessibility: A crowdsourcing approach through online reviews to inclusive urban design","authors":"Lingyao Li , Songhua Hu , Yinpei Dai , Min Deng , Parisa Momeni , Gabriel Laverghetta , Lizhou Fan , Zihui Ma , Xi Wang , Siyuan Ma , Jay Ligatti , Libby Hemphill","doi":"10.1016/j.compenvurbsys.2025.102329","DOIUrl":"10.1016/j.compenvurbsys.2025.102329","url":null,"abstract":"<div><div>As urban populations grow, the need for accessible urban design has become urgent. Traditional methods for assessing public perceptions of accessibility, such as surveys and interviews, are often resource-intensive and geographically limited in scope. Crowdsourcing via online reviews offers a valuable alternative to understanding public perceptions, and advancements in large language models (LLMs) can facilitate their use. In this study, we examine over one million Google Maps reviews from points of interests (POIs) across the United States and fine-tune the Llama 3 model using the Low-Rank Adaptation (LoRA) technique to identify public sentiment toward accessibility. At the POI level, most categories – restaurants, retail, hotels, and healthcare – show negative sentiments, indicating persistent barriers across key sectors. Socio-spatial regression analysis reveals that more positive sentiment is associated with areas that have higher proportions of white residents and greater socioeconomic advantage. Conversely, more negative sentiment is expressed in areas with higher concentrations of elderly and highly-educated populations. Interestingly, no clear link is found between the presence of disabilities and public sentiments, but a significant positive relationship does exist between disability-friendly scores and public perception. Overall, our findings demonstrate the value of crowdsourcing with LLM-enhanced analysis in identifying accessibility challenges and informing inclusive urban design, offering actionable insights for planners, policymakers, and advocates striving toward more equitable cities.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"122 ","pages":"Article 102329"},"PeriodicalIF":8.3,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144889251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ScaleFC: A scale-aware geographical flow clustering algorithm for heterogeneous origin-destination data ScaleFC：一种基于尺度感知的异构始发目的地数据地理流聚类算法

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-20 DOI: 10.1016/j.compenvurbsys.2025.102338

Huan Chen , Zhipeng Gui , Dehua Peng , Yuhang Liu , Yuncheng Ma , Huayi Wu

Exploring the cluster pattern of geographical flow facilitates the understanding of the underlying process of geographical phenomena among spatial locations. Despite recent advancements in identifying flow clusters, challenges remain when handling flow data with uneven length, heterogeneous density and weak connectivity. To solve the issues, this study proposes a Scale-aware Flow Clustering algorithm (ScaleFC). It identifies flow clusters of arbitrary lengths by employing an analytical scale to generate an adjustable neighborhood range of each flow. Meanwhile, inspired by the idea of boundary-seeking clustering, ScaleFC introduces partitioning flows to identify flow clusters with different densities, and separate the weakly-connected clusters. To validate the effectiveness, we compared ScaleFC with three mainstream baselines, i.e., AFC, FlowLF and FlowDBSCAN, on six synthetic datasets. The results presented that ScaleFC can accurately identify the clusters with complex structures, achieving an average accuracy improvement of 27 %, 17 %, and 15 % over the three competitors, respectively. The application on bike-sharing data with 16,140 flow pairs from Shanghai City demonstrated that ScaleFC is capable to capture both long-distance and short-distance movements, thereby providing a more comprehensive understanding to multi-scale human mobility patterns in geographical space.

探索地理流动的集群模式有助于理解空间区位间地理现象的内在过程。尽管最近在识别流簇方面取得了进展，但在处理长度不均匀、密度不均匀和连通性弱的流数据时仍然存在挑战。为了解决这一问题，本研究提出了一种规模感知流聚类算法（ScaleFC）。它通过采用分析尺度来生成每个流的可调邻域范围来识别任意长度的流簇。同时，受边界寻找聚类思想的启发，ScaleFC引入分区流来识别不同密度的流簇，并分离弱连接的簇。为了验证其有效性，我们在六个合成数据集上将ScaleFC与三个主流基线（即AFC， FlowLF和FlowDBSCAN）进行了比较。结果表明，ScaleFC可以准确地识别具有复杂结构的聚类，平均准确率比三个竞争对手分别提高27%，17%和15%。在上海市16140对共享单车流量数据上的应用表明，ScaleFC能够捕捉长距离和短距离的移动，从而更全面地了解地理空间中多尺度的人类移动模式。

{"title":"ScaleFC: A scale-aware geographical flow clustering algorithm for heterogeneous origin-destination data","authors":"Huan Chen , Zhipeng Gui , Dehua Peng , Yuhang Liu , Yuncheng Ma , Huayi Wu","doi":"10.1016/j.compenvurbsys.2025.102338","DOIUrl":"10.1016/j.compenvurbsys.2025.102338","url":null,"abstract":"<div><div>Exploring the cluster pattern of geographical flow facilitates the understanding of the underlying process of geographical phenomena among spatial locations. Despite recent advancements in identifying flow clusters, challenges remain when handling flow data with uneven length, heterogeneous density and weak connectivity. To solve the issues, this study proposes a Scale-aware Flow Clustering algorithm (ScaleFC). It identifies flow clusters of arbitrary lengths by employing an analytical scale to generate an adjustable neighborhood range of each flow. Meanwhile, inspired by the idea of boundary-seeking clustering, ScaleFC introduces partitioning flows to identify flow clusters with different densities, and separate the weakly-connected clusters. To validate the effectiveness, we compared ScaleFC with three mainstream baselines, i.e., AFC, FlowLF and FlowDBSCAN, on six synthetic datasets. The results presented that ScaleFC can accurately identify the clusters with complex structures, achieving an average accuracy improvement of 27 %, 17 %, and 15 % over the three competitors, respectively. The application on bike-sharing data with 16,140 flow pairs from Shanghai City demonstrated that ScaleFC is capable to capture both long-distance and short-distance movements, thereby providing a more comprehensive understanding to multi-scale human mobility patterns in geographical space.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"122 ","pages":"Article 102338"},"PeriodicalIF":8.3,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144863804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Digital planning for sustainable urban future 可持续城市未来的数字化规划

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-06 DOI: 10.1016/j.compenvurbsys.2025.102334

Yanliu Lin , Stan Geertman , Patrick Witte , Nuno Pinto

引用次数: 0

Wheelchair accessibility to public facilities via transits and analysis of delay factors—A case study of Shanghai, China 公共设施的轮椅可及性及其延迟因素分析——以上海为例

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-04 DOI: 10.1016/j.compenvurbsys.2025.102331

Luoan Yang , Wei Huang , Xintao Liu , Wanglin Yan

The pursuit of egalitarian and sustainable communities represents a collective aspiration and aligns with the United Nations’ Sustainable Development Goals. The public transit system, as the primary mode of mobility for wheelchair users in China, often imposes barriers that hinder travel or prolong travel times. It is essential to evaluate the spatial accessibility of public transit for wheelchair users to mitigate their social exclusion and enhance their participation within the community. However, there is a paucity of research on wheelchair transit accessibility and the factors contributing to prolonged travel times. This study introduces a wheelchair-accessible public transit route planning algorithm utilizing an online map API to acquire travel time and identify delay factors using the city of Shanghai as the study area, then evaluates spatial accessibility differences between wheelchair users and the general population in accessing public service facilities. Key findings include: (1) 73.9% of wheelchair transit routes encounter delays due to insufficient wheelchair facilities. (2) Parks show the largest accessibility gap, with wheelchair users’ accessibility at only 45% of that of the general population within the same time threshold. (3) Walking segment obstacles cause the longest delays, the most frequent delay factor is the lack of accessible metro station entrances, and SHAP values from the machine learning model furnish localized explanations regarding the impact of each delay factor. These findings reveal disparities in wheelchair transit accessibility and investigate factors causing delays, informing urban planning and infrastructure improvements in Shanghai and providing a reference for barrier-free development in other cities.

追求平等和可持续的社区是一种集体愿望，与联合国可持续发展目标相一致。公共交通系统作为中国轮椅使用者的主要出行方式，经常设置障碍，阻碍出行或延长出行时间。评估公共交通对轮椅使用者的空间可达性至关重要，以减轻他们的社会排斥，提高他们在社区中的参与度。然而，关于轮椅交通的可达性和导致出行时间延长的因素的研究却很少。本文以上海市为研究区，利用在线地图API，引入了一种轮椅无障碍公共交通路线规划算法，获取出行时间并识别延误因素，评估了轮椅使用者与一般人群在获取公共服务设施方面的空间可达性差异。主要发现包括：(1)73.9%的轮椅过境路线因轮椅设施不足而出现延误。(2)公园的可达性差距最大，在相同的时间阈值内，轮椅使用者的可达性仅为一般人群的45%。(3)步行段障碍物造成的延误时间最长，最常见的延误因素是缺乏可达的地铁站入口，机器学习模型的SHAP值对每个延误因素的影响提供了本地化的解释。研究结果揭示了上海轮椅交通可达性的差异，探讨了造成延误的因素，为上海的城市规划和基础设施改善提供了参考，并为其他城市的无障碍发展提供了参考。

{"title":"Wheelchair accessibility to public facilities via transits and analysis of delay factors—A case study of Shanghai, China","authors":"Luoan Yang , Wei Huang , Xintao Liu , Wanglin Yan","doi":"10.1016/j.compenvurbsys.2025.102331","DOIUrl":"10.1016/j.compenvurbsys.2025.102331","url":null,"abstract":"<div><div>The pursuit of egalitarian and sustainable communities represents a collective aspiration and aligns with the United Nations’ Sustainable Development Goals. The public transit system, as the primary mode of mobility for wheelchair users in China, often imposes barriers that hinder travel or prolong travel times. It is essential to evaluate the spatial accessibility of public transit for wheelchair users to mitigate their social exclusion and enhance their participation within the community. However, there is a paucity of research on wheelchair transit accessibility and the factors contributing to prolonged travel times. This study introduces a wheelchair-accessible public transit route planning algorithm utilizing an online map API to acquire travel time and identify delay factors using the city of Shanghai as the study area, then evaluates spatial accessibility differences between wheelchair users and the general population in accessing public service facilities. Key findings include: (1) 73.9% of wheelchair transit routes encounter delays due to insufficient wheelchair facilities. (2) Parks show the largest accessibility gap, with wheelchair users’ accessibility at only 45% of that of the general population within the same time threshold. (3) Walking segment obstacles cause the longest delays, the most frequent delay factor is the lack of accessible metro station entrances, and SHAP values from the machine learning model furnish localized explanations regarding the impact of each delay factor. These findings reveal disparities in wheelchair transit accessibility and investigate factors causing delays, informing urban planning and infrastructure improvements in Shanghai and providing a reference for barrier-free development in other cities.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"121 ","pages":"Article 102331"},"PeriodicalIF":8.3,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144766830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep contrastive learning for feature alignment: Insights from housing-household relationship inference 特征对齐的深度对比学习：来自住房-家庭关系推断的见解

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-03 DOI: 10.1016/j.compenvurbsys.2025.102328

Xiao Qian, Shangjia Dong, Rachel Davidson

Housing and household characteristics are key determinants of social and economic well-being, yet our understanding of their interrelationships remains limited. This study addresses this knowledge gap by developing a deep contrastive learning (DCL) model to infer housing-household relationships using the American Community Survey (ACS) Public Use Microdata Sample (PUMS). More broadly, the proposed model is suitable for a class of problems where the goal is to learn joint relationships between two distinct entities without explicitly labeled ground truth data. Our proposed dual-encoder DCL approach leverages co-occurrence patterns in PUMS and introduces a bisect K-means clustering method to overcome the absence of ground truth labels. The dual-encoder DCL architecture is designed to handle the semantic differences between housing (building) and household (people) features while mitigating noise introduced by clustering. To validate the model, we generate a synthetic ground truth dataset and conduct comprehensive evaluations. The model further demonstrates its superior performance in capturing housing-household relationships in Delaware compared to state-of-the-art methods. A transferability test in North Carolina confirms its generalizability across diverse sociodemographic and geographic contexts. Finally, the post-hoc explainable AI analysis using SHAP values reveals that tenure status and mortgage information play a more significant role in housing-household matching than traditionally emphasized factors such as the number of persons and rooms.

住房和家庭特征是社会和经济福祉的关键决定因素，但我们对其相互关系的理解仍然有限。本研究利用美国社区调查（ACS）公共使用微数据样本（PUMS）开发了一个深度对比学习（DCL）模型来推断住房-家庭关系，从而解决了这一知识差距。更广泛地说，所提出的模型适用于一类问题，其目标是学习两个不同实体之间的联合关系，而没有明确标记的基础真值数据。我们提出的双编码器DCL方法利用了PUMS中的共现模式，并引入了一种二分k均值聚类方法来克服缺乏基础真值标签的问题。双编码器DCL架构旨在处理住房（建筑）和家庭（人）特征之间的语义差异，同时减轻聚类带来的噪声。为了验证模型，我们生成了一个合成的地面真实数据集并进行了全面的评估。与最先进的方法相比，该模型进一步证明了其在捕捉特拉华州住房与家庭关系方面的卓越表现。在北卡罗来纳州进行的一项可转移性测试证实了它在不同社会人口和地理背景下的普遍性。最后，使用SHAP值的事后可解释人工智能分析表明，在住房-家庭匹配中，使用权地位和抵押信息比传统上强调的因素（如人数和房间数量）发挥更重要的作用。

{"title":"Deep contrastive learning for feature alignment: Insights from housing-household relationship inference","authors":"Xiao Qian, Shangjia Dong, Rachel Davidson","doi":"10.1016/j.compenvurbsys.2025.102328","DOIUrl":"10.1016/j.compenvurbsys.2025.102328","url":null,"abstract":"<div><div>Housing and household characteristics are key determinants of social and economic well-being, yet our understanding of their interrelationships remains limited. This study addresses this knowledge gap by developing a deep contrastive learning (DCL) model to infer housing-household relationships using the American Community Survey (ACS) Public Use Microdata Sample (PUMS). More broadly, the proposed model is suitable for a class of problems where the goal is to learn joint relationships between two distinct entities without explicitly labeled ground truth data. Our proposed dual-encoder DCL approach leverages co-occurrence patterns in PUMS and introduces a bisect K-means clustering method to overcome the absence of ground truth labels. The dual-encoder DCL architecture is designed to handle the semantic differences between housing (building) and household (people) features while mitigating noise introduced by clustering. To validate the model, we generate a synthetic ground truth dataset and conduct comprehensive evaluations. The model further demonstrates its superior performance in capturing housing-household relationships in Delaware compared to state-of-the-art methods. A transferability test in North Carolina confirms its generalizability across diverse sociodemographic and geographic contexts. Finally, the post-hoc explainable AI analysis using SHAP values reveals that tenure status and mortgage information play a more significant role in housing-household matching than traditionally emphasized factors such as the number of persons and rooms.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"121 ","pages":"Article 102328"},"PeriodicalIF":8.3,"publicationDate":"2025-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144764047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Urban planning in the age of large language models: Assessing OpenAI o1's performance and capabilities across 556 tasks 大型语言模型时代的城市规划：评估OpenAI 01在556个任务中的性能和能力

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-01 DOI: 10.1016/j.compenvurbsys.2025.102332

Xukai Zhao , He Huang , Tao Yang , Yuxing Lu , Lu Zhang , Ruoyu Wang , Zhengliang Liu , Tianyang Zhong , Tianming Liu

Integrating Large Language Models (LLMs) into urban planning presents significant opportunities to enhance efficiency and support data-driven city development strategies. Despite their potential, the specific capabilities of LLMs within the urban planning context remain underexplored, and the field lacks standardized benchmarks for systematic evaluation. This study presents the first comprehensive evaluation focused on OpenAI o1's performance and capabilities in urban planning, systematically benchmarking it against GPT-3.5 and GPT-4o using an original open-source benchmark comprising 556 tasks across five critical categories: urban planning documentation, examinations, routine data analysis, AI algorithm support, and thesis writing. Through rigorous testing and manual analysis of 170,627 words of generated output, OpenAI o1 consistently outperformed its counterparts, achieving an average performance score of 84.08 % compared to 69.30 % for GPT-4o and 45.27 % for GPT-3.5. Our findings highlight o1's strengths in domain knowledge mastery, basic operational competence, and coding capabilities, demonstrating its potential applications in information retrieval, urban data analytics, planning decision support, educational assistance, and LLM-based agent development. However, significant limitations were identified, including inability in urban design, susceptibility to fabricating information, moderate academic writing quality, challenges in high-level professional examinations, and spatial reasoning, and limited support for specialized or emerging AI algorithms. Future optimizations should prioritize enhancing multimodal integration, implementing robust validation mechanisms, adopting adaptive learning strategies, and enabling domain-specific fine-tuning to meet urban planners' specialized needs. These advancements would enable LLMs to better support the evolving demands of urban planning, allowing professionals to focus more on strategic decision-making and the creative aspects of the field.

将大型语言模型（llm）整合到城市规划中，为提高效率和支持数据驱动的城市发展战略提供了重要机会。尽管具有潜力，法学硕士在城市规划背景下的具体能力仍未得到充分探索，该领域缺乏系统评估的标准化基准。本研究首次对OpenAI o1在城市规划中的性能和能力进行了全面评估，使用原始的开源基准对其进行了系统的基准测试，该基准测试包括五个关键类别的556个任务：城市规划文档、考试、常规数据分析、人工智能算法支持和论文写作。通过对生成输出的170,627个单词的严格测试和人工分析，OpenAI 01始终优于同类产品，平均性能得分为84.08%，而gpt - 40和GPT-3.5的平均性能得分分别为69.30%和45.27%。我们的研究结果突出了o1在领域知识掌握、基本操作能力和编码能力方面的优势，展示了它在信息检索、城市数据分析、规划决策支持、教育辅助和基于llm的代理开发方面的潜在应用。然而，我们发现了显著的局限性，包括城市设计能力不足、易受虚假信息的影响、学术写作质量不高、在高水平专业考试和空间推理方面面临挑战，以及对专业或新兴人工智能算法的支持有限。未来的优化应优先考虑增强多模态集成，实现稳健的验证机制，采用自适应学习策略，并使特定领域的微调能够满足城市规划者的专门需求。这些进步将使法学硕士能够更好地支持城市规划不断变化的需求，使专业人员能够更多地关注该领域的战略决策和创造性方面。

{"title":"Urban planning in the age of large language models: Assessing OpenAI o1's performance and capabilities across 556 tasks","authors":"Xukai Zhao , He Huang , Tao Yang , Yuxing Lu , Lu Zhang , Ruoyu Wang , Zhengliang Liu , Tianyang Zhong , Tianming Liu","doi":"10.1016/j.compenvurbsys.2025.102332","DOIUrl":"10.1016/j.compenvurbsys.2025.102332","url":null,"abstract":"<div><div>Integrating Large Language Models (LLMs) into urban planning presents significant opportunities to enhance efficiency and support data-driven city development strategies. Despite their potential, the specific capabilities of LLMs within the urban planning context remain underexplored, and the field lacks standardized benchmarks for systematic evaluation. This study presents the first comprehensive evaluation focused on OpenAI o1's performance and capabilities in urban planning, systematically benchmarking it against GPT-3.5 and GPT-4o using an original open-source benchmark comprising 556 tasks across five critical categories: urban planning documentation, examinations, routine data analysis, AI algorithm support, and thesis writing. Through rigorous testing and manual analysis of 170,627 words of generated output, OpenAI o1 consistently outperformed its counterparts, achieving an average performance score of 84.08 % compared to 69.30 % for GPT-4o and 45.27 % for GPT-3.5. Our findings highlight o1's strengths in domain knowledge mastery, basic operational competence, and coding capabilities, demonstrating its potential applications in information retrieval, urban data analytics, planning decision support, educational assistance, and LLM-based agent development. However, significant limitations were identified, including inability in urban design, susceptibility to fabricating information, moderate academic writing quality, challenges in high-level professional examinations, and spatial reasoning, and limited support for specialized or emerging AI algorithms. Future optimizations should prioritize enhancing multimodal integration, implementing robust validation mechanisms, adopting adaptive learning strategies, and enabling domain-specific fine-tuning to meet urban planners' specialized needs. These advancements would enable LLMs to better support the evolving demands of urban planning, allowing professionals to focus more on strategic decision-making and the creative aspects of the field.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"121 ","pages":"Article 102332"},"PeriodicalIF":8.3,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144748888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analyzing public response to wildfires: A socio-spatial study using SIR models and NLP techniques 分析公众对野火的反应：使用SIR模型和NLP技术的社会空间研究

IF 8.3 1区地球科学 Q1 ENVIRONMENTAL STUDIES

Computers Environment and Urban Systems

Pub Date : 2025-08-01 DOI: 10.1016/j.compenvurbsys.2025.102333

Zihui Ma , Guangxiao Hu , Ting-Syuan Lin , Lingyao Li , Songhua Hu , Loni Hagen , Gregory B. Baecher

The increasing frequency and severity of wildfires pose significant risks to communities, infrastructure, and the environment, especially in Wildland-Urban Interface (WUI) areas. Effective disaster management requires understanding how the public perceives and responds to wildfire threats in near-real-time. This study uses social media data to assess public responses (including collective perceptions/reactions) and explores how these responses are linked to city-level community characteristics. Specifically, we leveraged a transformer-based topic modeling technique called BERTopic to identify wildfire response-related topics and then utilized the Susceptible-Infectious-Recovered (SIR) model to compute two key metrics — public response awareness and resilience indicators. Additionally, we used GIS-based spatial analysis to map wildfire responses and the relationships with four groups of city-level factors (racial/ethnic, socioeconomic, demographic, and wildfire-specific). Our findings reveal significant geographic and socio-spatial differences in public responses. Southern California cities with larger Hispanic populations demonstrate higher wildfire awareness and resilience. In contrast, urbanized regions in Central and Northern California exhibit lower awareness levels. Furthermore, resilience is negatively correlated with unemployment rates, particularly in southern regions where higher unemployment aligns with reduced resilience. These findings highlight the need for targeted and equitable wildfire management strategies to improve the adaptive capacity of WUI communities.

野火发生的频率和严重程度日益增加，对社区、基础设施和环境构成了重大风险，特别是在荒地-城市交界地区。有效的灾害管理需要了解公众如何近乎实时地感知和应对野火威胁。本研究使用社交媒体数据来评估公众的反应（包括集体的看法/反应），并探讨这些反应如何与城市层面的社区特征联系起来。具体来说，我们利用一种名为BERTopic的基于变压器的主题建模技术来识别与野火响应相关的主题，然后利用易感-感染-恢复（SIR）模型来计算两个关键指标——公众响应意识和恢复指标。此外，我们使用基于gis的空间分析来绘制野火响应及其与四组城市层面因素（种族/民族、社会经济、人口统计学和野火特异性）的关系。我们的研究结果揭示了公众反应的显著地理和社会空间差异。西班牙裔人口较多的南加州城市表现出更高的野火意识和恢复能力。相比之下，加州中部和北部的城市化地区表现出较低的意识水平。此外，弹性与失业率呈负相关，特别是在南部地区，高失业率与弹性降低相一致。这些发现强调需要有针对性和公平的野火管理战略，以提高WUI社区的适应能力。

{"title":"Analyzing public response to wildfires: A socio-spatial study using SIR models and NLP techniques","authors":"Zihui Ma , Guangxiao Hu , Ting-Syuan Lin , Lingyao Li , Songhua Hu , Loni Hagen , Gregory B. Baecher","doi":"10.1016/j.compenvurbsys.2025.102333","DOIUrl":"10.1016/j.compenvurbsys.2025.102333","url":null,"abstract":"<div><div>The increasing frequency and severity of wildfires pose significant risks to communities, infrastructure, and the environment, especially in Wildland-Urban Interface (WUI) areas. Effective disaster management requires understanding how the public perceives and responds to wildfire threats in near-real-time. This study uses social media data to assess public responses (including collective perceptions/reactions) and explores how these responses are linked to city-level community characteristics. Specifically, we leveraged a transformer-based topic modeling technique called BERTopic to identify wildfire response-related topics and then utilized the Susceptible-Infectious-Recovered (SIR) model to compute two key metrics — public response awareness and resilience indicators. Additionally, we used GIS-based spatial analysis to map wildfire responses and the relationships with four groups of city-level factors (racial/ethnic, socioeconomic, demographic, and wildfire-specific). Our findings reveal significant geographic and socio-spatial differences in public responses. Southern California cities with larger Hispanic populations demonstrate higher wildfire awareness and resilience. In contrast, urbanized regions in Central and Northern California exhibit lower awareness levels. Furthermore, resilience is negatively correlated with unemployment rates, particularly in southern regions where higher unemployment aligns with reduced resilience. These findings highlight the need for targeted and equitable wildfire management strategies to improve the adaptive capacity of WUI communities.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"121 ","pages":"Article 102333"},"PeriodicalIF":8.3,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144748887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0