Date of Award
Civil and Environmental Engineering
Collision risk, Crash potential index, Heavy vehicle, Lane-change collision, Rear-end collision, Surrogate safety measure
CC BY-NC-ND 4.0
This study analyzes car-following and lane-change conflicts in car-heavy vehicle mixed traffic flow on freeways using three surrogate safety measures - time-to-collision (TTC), post-encroachment-time (PET) and crash potential index (CPI). The surrogate safety measures were estimated for different types of lead and following vehicles (car or heavy vehicle) using the individual vehicle trajectory data. The data were collected from a segment of the US-101 freeway in Los Angeles, California, U.S.A. For car-following conflicts, the distributions of TTC and PET were significantly different among different types of lead and following vehicles. For lane-change conflicts between the lane-change vehicle and the trailing vehicle in the target lane, CPIs were higher for angle conflicts than rear-end conflicts. It was also found that the CPI was generally higher for a given spacing interval when the following vehicle is a heavy vehicle in both car-following and lane-change conflicts. This indicates that heavy vehicle’s lower braking capability significantly increases collision risk. This study also validates the CPI using historical crash data and the loop detector data extracted a few minutes before crash time upstream and downstream of crash locations. The data were obtained from a section of the Gardiner Expressway, Ontario, Canada. The result shows that the values of CPI were consistently higher for the crash case than the non-crash case. This shows that CPI can be used to capture the collision risk during car-following and lane-change maneuver on freeways. The findings suggest that the differences in collision risk among different vehicle pair types should be considered in the assessment of safety of car-heavy vehicle mixed traffic flow.
Zhao, Peibo, "Safety Evaluation of Car-Truck Mixed Traffic Flow on Freeways Using Surrogate Safety Measures" (2016). Electronic Theses and Dissertations. 5923.