Analysis of factors influencing crash injury severities at highway–rail grade crossings accommodating for unobserved heterogeneity

Abstract

Introduction: This research aims to identify and understand the risk factors associated with injury severities in accidents occurring at highway–rail grade crossings in the context of the developing country, Thailand. Method: The mixed logit model was employed by analyzing crash data over 10 years, from 2012 to 2022. Results: The analysis revealed a number of significant factors associated with severe or fatal crashes. These included accidents that occurred between midnight and 6 a.m., collisions involving pickup cars or heavy trucks, roads with a “no overtaking zone” sign, intersections classified as Type B1 (defined by the presence of only warning posts and horizontal crossing barriers), and intersections without adequate traffic control devices. In contrast, hazard markers on pavements and traverse rumble strips were positively correlated with property damage only crashes, where no injuries were reported. Discussion: This study provides insights into contributory factors to accidents at highway–rail grade crossing. Based on these key findings, the study recommends increasing nighttime visibility at railway grade crossings, developing targeted education and training programs for pickup car and truck drivers, installing hazard markers and traverse rumble strips, and considering physical barriers, such as bollards or delineators, to discourage overtaking maneuvers near the railway crossing.