Abstract: Determining crash causation has always been a focus and a difficulty in the field of traffic safety. Previous research has had to rely on insufficient crash data and crash causation analysis methods limited to a single crash, and has not taken advantage of the application value of pre-crash scenarios in causation analysis. This study therefore proposed a two-stage crash causation analysis method based on pre-crash scenarios, and analyzed crashes and near crashes (CNCs) using naturalistic driving data. From the Shanghai Naturalistic Driving Study (SH-NDS), 572 CNCs were extracted, and 25 pre-crash scenarios were identified using the Pre-Crash Scenario Typology. In-depth investigations of CNCs in the same scenario were analyzed to determine the causes of crashes using the proposed systematic crash causation derivation framework, which summarizes the causation patterns in each scenario based on the interaction of humans, vehicles, infrastructure, and environment subsystems. The differences between the causation patterns of three common pre-crash scenarios (rear-end, lane change and pedalcyclist collisions) were determined through statistical analysis. Following too closely and non-driving-related distraction were important causes of rear-end scenarios. Distraction, as well as willful behavior and violation of traffic laws was a common pattern (61.2%) in lane change pre-crash scenarios. Pedalcyclist scenarios leading to CNCs were particularly impacted by pedalcyclists violating traffic regulations, visual obstructions, and inadequate lanes for non-motorized vehicles. Based on causation patterns, this study suggests countermeasures for the three scenario types. These findings provide support for safety improvement projects and the development of advanced driver assistance systems.
Xuesong Wang*, Qian Liu, Feng Guo, Shouen Fang, Xiaoyan Xu. Crashes and Near Crashes Causation Analysis Using Naturalistic Driving Data.Transportation Research Board 101th Annual Meeting, Washington D.C., USA, 2022. 1.9-13.