A Spatial and Data-Driven Analysis of Factors Affecting Pedestrian Safety in High-Traffic Areas of Mashhad Metropolis

A Spatial and Data-Driven Analysis of Factors Affecting Pedestrian Safety in High-Traffic Areas of Mashhad Metropolis

Mahdi Ghandehari1 toktam hanaee2

1) 1.PhD Researcher in Urban Planning, Faculty of Art and Architecture, Islamic Azad University, Mashhad Branch;
2) Associate Professor, Department of Urban Planning, Faculty of Art and Architecture, Islamic Azad University, Mashhad Branch;

IOI Code : AAHF-ABKZCC
Publication : National Conference on Architecture and Art Technologies, Smart City, Soft and Creative Industries and Technological Culture(icact.ir)
Abstract :
    Pedestrian safety has become one of the key challenges of sustainable urban planning and human-oriented transportation, particularly in large metropolitan and pilgrimage cities. In Mashhad, as one of the most important pilgrimage destinations in Iran, pedestrian safety is strongly affected by high population density, temporal fluctuations of visitors and pilgrims, intensive land-use activities, and increasing pressure on the pedestrian access network. Accordingly, this study aims to model pedestrian access network planning and analyze the factors affecting pedestrian safety in the metropolis of Mashhad with an emphasis on big data. This research is applied in terms of purpose and analytical-modeling in terms of method. The study uses various data sources, including pedestrian crash data, street network characteristics, land-use indicators, traffic-related variables, environmental attributes, and temporal indicators associated with population movement patterns. Spatial analysis techniques were applied to identify unsafe pedestrian patterns and high-risk clusters. In addition, Geographically Weighted Regression was used to examine the spatial heterogeneity of influencing factors, while the Random Forest algorithm was employed to identify important variables and predict high-risk areas.The findings indicate that pedestrian safety risks in Mashhad are not randomly distributed but are spatially clustered, particularly in central, pilgrimage-oriented, highly active, and traffic-intensive areas. The modeling results reveal that variables such as intersection density, traffic volume, vehicle speed, pedestrian network continuity, sidewalk quality, land-use attraction, proximity to public transport stations, street width, and temporal population fluctuations significantly influence pedestrian safety. Moreover, the effects of these variables vary across different urban zones, confirming the spatially heterogeneous nature of pedestrian safety risks. The study concludes that uniform and generalized approaches to pedestrian safety management are insufficient for a complex urban context such as Mashhad. Instead, data-driven, place-based, and context-sensitive strategies are required. The application of big data and predictive modeling can help identify high-risk areas before severe crashes occur and support targeted interventions. The study recommends improving pedestrian network connectivity, redesigning unsafe intersections, implementing traffic calming measures, managing pedestrian and vehicle flows intelligently, prioritizing pilgrimage and high-density areas, and adopting complete streets policies to enhance pedestrian safety in Mashhad.

Keywords : pedestrian safety spatial analysis geographically weighted regression random forest built environment mashhad sustainable transport