Dynamic spatiotemporal graph network for traffic accident risk prediction

Traffic accidents remain major public safety concerns, often causing severe injuries, deaths, and economic costs, especially in rapidly urbanizing areas. Accurate traffic accident risk prediction is crucial for developing effective strategies to reduce accidents and enhance urban mobility. However,...

Full description

Saved in:
Bibliographic Details
Main Authors: Pengcheng Zhang, Wen Yi, Yongze Song, Penggao Yan, Peng Wu, Ammar Shemery, Keith Hampson, Albert P. C. Chan
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:GIScience & Remote Sensing
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/15481603.2025.2514330
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849334649480806400
author Pengcheng Zhang
Wen Yi
Yongze Song
Penggao Yan
Peng Wu
Ammar Shemery
Keith Hampson
Albert P. C. Chan
author_facet Pengcheng Zhang
Wen Yi
Yongze Song
Penggao Yan
Peng Wu
Ammar Shemery
Keith Hampson
Albert P. C. Chan
author_sort Pengcheng Zhang
collection DOAJ
description Traffic accidents remain major public safety concerns, often causing severe injuries, deaths, and economic costs, especially in rapidly urbanizing areas. Accurate traffic accident risk prediction is crucial for developing effective strategies to reduce accidents and enhance urban mobility. However, predicting traffic accident risks is challenging due to the relationships among factors such as weather, traffic conditions, and road characteristics, along with capturing spatial correlations of traffic accidents across different time scales. To address these challenges, we propose the dynamic spatial-temporal accident risk network (DSTAR-Net). Our model uses channel-wise convolutional neural networks to detect spatial accident patterns across weekly, daily, and hourly time scales with automatic weight learning, simultaneously employing graph convolutional networks to process road network features, population feature while integrating external data like weather and dates. The dynamic learning of spatial correlations, combined with the integration of road characteristics and contextual variables, significantly enhances the accuracy of traffic accident predictions. Experiments in Perth show the DSTAR-Net outperforms state-of-the-art models with RMSE 24.901, Recall 21.59%, and MAP 0.0721. Notably, the weights learned by our model indicate that hourly patterns have the highest weight at 0.390, while weekly trends carry the lowest weight at 0.255, suggesting that recent traffic conditions have the most significant influence on accident risks. This study provides a foundational framework for predicting traffic accident risks, aiding urban planners and policymakers in enhancing road safety and traffic management in cities.
format Article
id doaj-art-d84f85fdddec4bd48e77c3a03c3683f0
institution Kabale University
issn 1548-1603
1943-7226
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series GIScience & Remote Sensing
spelling doaj-art-d84f85fdddec4bd48e77c3a03c3683f02025-08-20T03:45:31ZengTaylor & Francis GroupGIScience & Remote Sensing1548-16031943-72262025-12-0162110.1080/15481603.2025.2514330Dynamic spatiotemporal graph network for traffic accident risk predictionPengcheng Zhang0Wen Yi1Yongze Song2Penggao Yan3Peng Wu4Ammar Shemery5Keith Hampson6Albert P. C. Chan7Department of Building and Real Estate, The Hong Kong Polytechnic University, Hong Kong, ChinaDepartment of Building and Real Estate, The Hong Kong Polytechnic University, Hong Kong, ChinaSchool of Design and the Built Environment, Curtin University, Perth, AustraliaDepartment of Aeronautical and Aviation Engineering, Faculty of Engineering, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, ChinaSchool of Design and the Built Environment, Curtin University, Perth, AustraliaSustainable Built Environment National Research Centre, Curtin University, Perth, AustraliaSustainable Built Environment National Research Centre, Curtin University, Perth, AustraliaDepartment of Building and Real Estate, The Hong Kong Polytechnic University, Hong Kong, ChinaTraffic accidents remain major public safety concerns, often causing severe injuries, deaths, and economic costs, especially in rapidly urbanizing areas. Accurate traffic accident risk prediction is crucial for developing effective strategies to reduce accidents and enhance urban mobility. However, predicting traffic accident risks is challenging due to the relationships among factors such as weather, traffic conditions, and road characteristics, along with capturing spatial correlations of traffic accidents across different time scales. To address these challenges, we propose the dynamic spatial-temporal accident risk network (DSTAR-Net). Our model uses channel-wise convolutional neural networks to detect spatial accident patterns across weekly, daily, and hourly time scales with automatic weight learning, simultaneously employing graph convolutional networks to process road network features, population feature while integrating external data like weather and dates. The dynamic learning of spatial correlations, combined with the integration of road characteristics and contextual variables, significantly enhances the accuracy of traffic accident predictions. Experiments in Perth show the DSTAR-Net outperforms state-of-the-art models with RMSE 24.901, Recall 21.59%, and MAP 0.0721. Notably, the weights learned by our model indicate that hourly patterns have the highest weight at 0.390, while weekly trends carry the lowest weight at 0.255, suggesting that recent traffic conditions have the most significant influence on accident risks. This study provides a foundational framework for predicting traffic accident risks, aiding urban planners and policymakers in enhancing road safety and traffic management in cities.https://www.tandfonline.com/doi/10.1080/15481603.2025.2514330Traffic accidents predictionspatial-temporal analysisurban computingintelligent transportation
spellingShingle Pengcheng Zhang
Wen Yi
Yongze Song
Penggao Yan
Peng Wu
Ammar Shemery
Keith Hampson
Albert P. C. Chan
Dynamic spatiotemporal graph network for traffic accident risk prediction
GIScience & Remote Sensing
Traffic accidents prediction
spatial-temporal analysis
urban computing
intelligent transportation
title Dynamic spatiotemporal graph network for traffic accident risk prediction
title_full Dynamic spatiotemporal graph network for traffic accident risk prediction
title_fullStr Dynamic spatiotemporal graph network for traffic accident risk prediction
title_full_unstemmed Dynamic spatiotemporal graph network for traffic accident risk prediction
title_short Dynamic spatiotemporal graph network for traffic accident risk prediction
title_sort dynamic spatiotemporal graph network for traffic accident risk prediction
topic Traffic accidents prediction
spatial-temporal analysis
urban computing
intelligent transportation
url https://www.tandfonline.com/doi/10.1080/15481603.2025.2514330
work_keys_str_mv AT pengchengzhang dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT wenyi dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT yongzesong dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT penggaoyan dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT pengwu dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT ammarshemery dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT keithhampson dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction
AT albertpcchan dynamicspatiotemporalgraphnetworkfortrafficaccidentriskprediction