Exploring the potential of disaggregated connected vehicle probe data in freeway congestion analysis: Insights from the I-20/59 freeway

Traditional traffic data sources, such as sensors (including radar, Bluetooth, and inductive loops) and aggregated probe vehicle data, provide insights but remain constrained by limited coverage, fragmentation, and low granularity. This study uses disaggregated connected vehicle probe (DCVP) data to provide high-resolution insights into congestion dynamics. Advanced trajectory alignment and clustering techniques identify congestion clusters, analyze queue lengths, and examine shockwave propagation. Driver behavior analysis, such as sudden deceleration and illegal turns, reveal safety risks and secondary crash potential. A comparative analysis of freeway and alternate routes evaluates the impact of congestion on travel efficiency. Additionally, a congestion scoring framework quantifies severity, validated against regional incident data to detect disruptions not tied to reported incidents. The study observed queue lengths reaching up to 7 miles and recovery times as long as 6 hours during major crash events. Furthermore, 74 significant congestion events were detected exclusively through DCVP data, indicating persistent bottlenecks that are not reflected in regional incident records which highlight the value of DCVP data in diagnosing traffic patterns and enabling adaptive decision-making.