Exploring Road Traffic Accidents Hotspots Using Clustering Algorithms and GIS-Based Spatial Analysis

Abstract

This study conducts a comprehensive spatial analysis of road traffic accidents (RTAs) in Najran, a city emblematic of rapid urbanization in Saudi Arabia, which is facing significant public safety challenges due to an increase in vehicular traffic. By means of a dataset from 2022, we explore the spatial distribution of RTAs across the city’s districts by employing advanced clustering algorithms, including Density-based Spatial Clustering of Applications with Noise (DBSCAN) and Hierarchical Agglomerative Clustering (HAC), as well as GIS-based density analysis, proximity analysis, and spatial interpolation, to unveil accident hotspots and disparities in emergency service coverage. Our findings reveal that (1) the HAC model, based on the Silhouette and Calinski-Harabasz Scores, performs better in identifying accident hotspots; (2) significant concentrations of accidents are observed along major highways and arterial roads, pinpointing critical hotspots within the city’s fabric; (3) proximity analysis indicates gaps in the coverage of ambulance services and public hospitals relative to high-incident areas; (4) through spatial interpolation, detailed visualizations of RTA distributions are provided, revealing diverse accident patterns across Najran. The study highlights the critical role of spatial analysis in identifying high-risk areas and provides valuable insights for transport planners and public safety officials, supporting the development of targeted strategies to improve road safety and enhance emergency service responses.