Injury severity of truck-involved crashes in work zones on rural and urban highways: Accounting for unobserved heterogeneity

Abstract

Truck-related crashes in the work zones always resulted in severe injury outcomes. This paper analyses the injury severity of truck-involved crashes in the work zones based on a ten-year (2005–2014) data collected in North Carolina. A wide range of variables is considered, including the driver, environmental, work-zone, road, and vehicle characteristics. The mixed logit (MXL) and partial proportional odds (PPO) logit models are employed to investigate the potential unobserved heterogeneity associated with the indicator of urban and rural highways. A comparison of the two models shows that PPO outperforms the MXL model. The estimation results showed that contributing factors resulting in injury-severity levels of work zones crashes involving trucks on rural and urban highways are significantly different. Elasticity effects are provided and used to evaluate the significance of the variables in the rural and urban models. The findings of this study help understand the injury-severity outcomes in truck-involved crashes in the rural and urban work zones.