Injury severity analysis in vehicle-pedestrian crashes

Abstract

Since traffic accident data has been typically collected by the police who record accident information based on estimation through the engineering methods and statements of accident-involved people and witnesses in the vicinity of an accident, accident data recorded by the police may inevitably include errors. However, the technological progression of in-vehicle driving recording devices such as event data recorder, vehicle black box (VBB), and various sensors using in naturalistic driving study tackles the change of methodological paradigms in traffic safety research. The objective of this study is to analyze injury severity in taxi-pedestrian crashes using more accurate crash data by VBBs such as time-to-collision, crash speed, crash angle, and crash region of vehicle and pedestrian. To carry out this study, VBB data for two years (2010-2011) collected by taxies operating in Incheon, Korea are used. Then, the ordered probit model is applied to analyze injury severity in taxi-pedestrian crashes.