Crash injury severity analysis of E-Bike Riders: A random parameters generalized ordered probit model with heterogeneity in means

Abstract

Electric bike (E-bike) fatal crashes have increased by 34% during the period between 2014 and 2016, raising a great challenge for traffic safety in China. This study examines the effects of road traffic characteristics, environmental factors, crash characteristics and rider demographic factors on e-bike riders' injury severity. A total of 2222 police-reported crash records of e-bike riders in a representative developing area in China-Hunan province from 2014 to 2016 is used for the current study. To account for the ordinal nature of crash severity and to incorporate unobserved heterogeneity at the observation level, a random parameters generalized ordered probit model with heterogeneity in means (RGOP-HM) is applied for the injury severity analysis. For examining the efficiency of the proposed model in modeling e-bike rider injury severity, ordered probit models, random parameters ordered probit models, random parameters generalized ordered probit models were also estimated. The superiority of RGOP-HM in terms of model fitness statistics indicates the importance of relaxing the limitations of traditional ordinal probability methods. The results of RGOP-HM revealed a wide range of factors associated with the e-bike injuries, including horizontal curves, roads with a high posted speed limit, traffic sign-controlled intersections, dim light, unlighted darkness, single-vehicle crashes, collisions with a heavy motorized vehicle, rider age over 44 (45–59, above 59), and rural areas. Based on the factors contributing to the increased injury severity, several safety implications are proposed from the perspective of engineering, education, and enforcement (3E). This study's findings could provide references for the development of targeted countermeasures to improve e-bike traffic safety in China.