Comprehensive Safety Diagnosis Using Automated Video Analysis: Applications to an Urban Intersection in Edmonton,Alberta,Canada

This study conducted an automated safety diagnosis for a major signalized intersection in the city of Edmonton, Alberta, Canada. The study was motivated by concerns raised about the potential safety implications of increasing the speed limit across the major road corridor of the intersection to 60 km/h from the current posted speed of 50 km/h. The diagnosis was performed with video data collected at the intersection for two consecutive days. Traffic conflicts at the location were identified, analyzed, and categorized according to such criteria as severity and road user type. Temporal and spatial violations were automatically identified. Automated data collection for traffic counts and travel speeds was performed and validated. It was observed that the high frequency of conflicts between vehicles and vulnerable road users (i.e., pedestrians and cyclists) and the presence of heavy vehicles could lead to more severe conflicts and possible collisions if the speed limit is raised. According to the outcome of the performed analysis, it was recommended that the existing speed limit not be raised, but kept at the current level. Several safety countermeasures were suggested to improve the safety at the intersection. This study demonstrated the practical application of automated traffic conflict analysis technology and its ability to help traffic engineers conduct a comprehensive safety assessment.