A Latent Segmentation Generalized Ordered Logit Model to Exa mine Factors Influencing Driver Injury Severity

摘要

This paper formulates and estimates an econometric model, referred to as the latent segmentationbased ordered logit (LSGOL) model, for examining driver injury severity. The proposed modelprobabilistically segments drivers (involved in a crash) into different injury risk segments basedon crash characteristics to recognize that the impacts of exogenous variables on driver injuryseverity level can vary across drivers based on both observed and unobserved crashcharacteristics. The proposed model is estimated using data drawn from Victoria Crash Databasefrom Australia for the years 2006 through 2010. The model estimation incorporates the influenceof a comprehensive set of exogenous variables grouped into six broad categories: crashcharacteristics, driver characteristics, vehicle characteristics, roadway design attributes,environmental factors and situational factors. The results clearly highlight the need forsegmentation based on crash characteristics. The crash characteristics that affect the allocation ofdrivers into segments include: collision object, trajectory of vehicle’s motion and manner ofcollision. Further, the key factors resulting in severe driver injury severity are driver age 65 andabove, driver ejection, not wearing seat belts, and collision in a high speed zone. The factorsreducing injury severity include presence of pedestrian control, presence of roundabout, driving apanel van, unpaved road condition and presence of passenger.