Self-Organized Neural Network Method to Identify Crash Hotspots

Crash hotspot identification (HSID) is an essential component of traffic management authorities’ efforts to improve safety and allocate limited resources. This paper presents a method for identifying hotspots using self-organizing maps (SOM). The SOM was used identify high-risk areas based on five commonly HSID methods: crash frequency, equivalent property damage only, rate, empirical Bayes, the societal risk-based method. Crashes major road in Iran were examined proposed Based these criteria, locations grouped into six clusters, which provided appropriate criteria each location depending importance cluster. findings show that tends focus with more crashes deaths, demonstrating research methodology appropriate.