Resheff, Yehezkel (University of Toronto), Sher, Mali (Holon Institiute of Technology), Adler, Nicole (Hebrew Univeristy)

The Temporal Dynamics of Road Traffic Crash Hotspots

Scheduled for presentation during the Poster Session "Traffic prediction and estimation III" (ThAT13), Thursday, September 26, 2024, 10:30−12:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

Despite tremendous efforts on multiple fronts, road traffic crashes remain a major ongoing cause of preventable loss of life and serious injury. Many policy and intervention efforts aim to reduce the number and severity of collisions. Of these, the most direct is the identification and correction of locations on the road network that are hotspots of severe crashes. So far, relatively limited attempts have been made to characterize the temporal dynamics of hotspot activity beyond counts in spatio-temporal bins. In this paper we propose a framework for analysis of hotspot dynamics in terms of the sequence of years in which each hotspot was active, based on a spatial tracking algorithm and a taxonomy of temporal patterns. We apply the method to a large nationwide dataset spanning 14 years and consisting of over 100,000 urban crashes in Israel. Results show that the proposed framework is able to track hotspots as they evolve over years, and detect distinct temporal structures in the sequence of years in which hotspots are active. This information has the potential to inform intervention strategies and the prioritization of the most promising hotspot locations for intervention.