Leveraging Connected Vehicle Trajectories to Estimate Driver Compliance at Stop-Controlled Intersections: A Novel Approach

Abstract

The availability of commercial connected vehicle (CV) data has significantly enhanced traffic safety research by facilitating the analysis of vehicle trajectories for safety applications. Stop sign intersections are critical areas with a high risk of crashes if drivers fail to comply. Estimating compliance rates at these intersections using traditional sensors is challenging and costly, making CV data an effective alternative. This study presents a novel approach using CV trajectories to estimate compliance rates at stop-controlled intersections. The method was applied to 4,365 intersections in Iowa, U.S., including one-way, two-way, and all-way stop configurations in both rural and urban areas. The study used 1 year of CV waypoint data from April, 2022, to March, 2023, with market penetration rates ranging from 4% to 10%, to calculate compliance rates for various movements and time settings, including different times of day and seasonal variations. Results showed that, on average, rural intersections had 15% lower compliance rates than urban intersections. No significant seasonal variations were observed, but compliance was generally lower at night. Further analysis using statistical techniques found no spatial dependency at the county level. A binomial regression on the selected attributes has identified location, intersection type, and turning movements as significant factors. This method can be used to monitor 95% of the intersections in Iowa. The generated compliance rates can be used for targeted improvements by transportation agencies for intersection safety and smart enforcement.