heavy vehicle crashes at unsignalized intersections: embracing unobserved heterogeneity in injury severity analysis

Studies on fatal crashes involving heavy vehicle crashes at unsignalized intersections typically emphasize the examination of injury severity at the individual level rather than analyzing the crash itself. the objective of this study is to bridge the existing gap in the literature regarding the injuries caused by heavy vehicles at unsignalized intersections. not only that but this study also seeks to find sound answers to the research questions: firstly, how does the absence of a traffic signal or any other form of traffic control affect the injury severity of heavy vehicle drivers? secondly, what are the factors that contribute to injury severity sustained by drivers of heavy vehicles at unsignalized intersections? this study utilized the data regarding heavy vehicle crashes that occurred at unsignalized intersections from 2007 to 2013 in the state of washington. a total of 4638 crashes involving heavy vehicles at unsignalized intersections were reported between 2007 and 2013. three main injury severity categories were considered: severe (fatal and incapacitating injuries), minor (non-incapacitating and possible injuries), and no injury. a mixed logit model is estimated to achieve the objectives of the study and also to tackle the issue of unobserved heterogeneity. the estimation results show that crashes on wet roadways, crashes involving left-turning movements, and crashes involving sober drivers produced obvious heterogeneous impacts on driver injury severity while the rest of the factors had homogeneous impacts. the findings indicate that sober drivers and the lack of restraining systems increased the odds of sustaining severe injuries by 0.0078 and 0.0022 as suggested by marginal effects. in contrast, no injury chances increased if crashes occurred in daylight by 0.0512. this study can aid in a deeper understanding of the risk factors of heavy vehicle crashes at unsignalized intersections so that safety at such locations can be improved by recommending effective countermeasures. © the author(s), under exclusive licence to the iran university of science and technology 2024.