Spatial analysis of road traffic accident hotspots: evaluation and validation of recent approaches using road safety audit

Abstract

Reducing road traffic accidents (RTA) and their socioeconomic costs is an increasingly important priority in many countries. In recent years, many authors have proposed various approaches to analyse historical RTA data using Geographic Information System (GIS) tools, so that locations with high instances or risk of RTA – i.e., RTA hotspots or blackspots – could be identified and ranked. This could then enable limited road engineering resources to be prioritised for remediating more critical RTA hotspots and therefore reducing RTA rates. This paper presents the development of a new validation method to evaluate the application of four different hotspot analysis methods in ESRI ArcGIS 10.2 to identify and rank RTA hotspots using historical RTA data on a section of a road in Brunei Darussalam as a case study. A road safety audit was undertaken for six locations along the study road section to identify the risk levels at these locations. These risk levels were then compared with risk levels at the same six locations computed by four hotspot analysis methods: Network KDE by SANET, KDE+, Getis-Ord Gi*, and a recently proposed risk-based method – STAA – that accounts for RTA frequency, severity and socioeconomic costs. It was found that SANET-KDE overestimated the risk levels of road sections that are free of RTA. The results from KDE + showed the most deviations compared to the identified risk levels by the road safety audit. Getis-Ord Gi* was not able to identify RTA hotspots along the study road section of Jalan Tutong. However, the STAA method provided the most consistent results in comparison with the identified risk levels by the road safety audit. STAA, by having a better validity compared to other methods, provides a promising way to prioritise RTA hotspots for remediation, and thus reducing the associated socioeconomic costs. In addition, these findings demonstrated that the novel method of validation via risk levels determined by road safety audit was able to provide an independent measure of the validity of the hotspot analysis (HSA) methods.

Keywords:

• GIS
• traffic safety
• hotspot analysis
• blackspot identification
• road traffic accidents

Acknowledgments

The authors are grateful to the Brunei Research Council and Universiti Teknologi Brunei for their funding support and to the Brunei National Road Safety Council and the Royal Brunei Police Force for providing the necessary road safety data and information. The authors also thank the Transport Research Centre (CDV) and the Centre for Spatial Information Science (CSiS) at the University of Tokyo for providing a licence to use KDE + toolbox and SANET toolbar, respectively. Further appreciation is extended to Safwanah Ni’matullah binti Mohd Said for her assistance with the Getis-Ord Gi* hotspot analysis method.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.