Differential analysis of factors influencing motorcycle single-vehicle crash severity on roadway segments and intersections.

Abstract

Objective: Motorcycle single-vehicle crashes occurring on roadway segments and intersections exhibit significant differences. This study aims to explore the differences in the key factors influencing the severity of motorcycle single-vehicle crashes at roadway segments and intersections, based on historical accident data from Shandong Province, China. Additionally, it examines the temporal stability of these determinants and the unobserved heterogeneity over time.

Methods: To account for heterogeneity in the data, Random parameters logit models with heterogeneity in means and variances are employed across different time periods and junction types. The analysis investigates the impact of various factors-driver, vehicle, road, and environmental-on the injury severity of motorcyclist. Additionally, the study assesses the temporal stability and unobserved heterogeneity of determinants in motorcycle crashes. Likelihood ratio tests are conducted to evaluate the transferability of model estimates across different years, and the marginal effects of explanatory variables are calculated. The results reveal significant variations in the effects of influencing factors between roadway segments and intersections.

Results: The driver age under 30 and during dawn or dusk are significantly associated with severe motorcycle single-vehicle crashes occurring exclusively at roadway segments. In contrast, visibility <100 m, visibility between 100 and 200 m, as well as the summer and winter seasons, are significant factors only for severe motorcycle single-vehicle crashes at intersections. Overall temporal instability is observed, while several indicators, including driver age over 50 and non-clear weather conditions, demonstrate relative spatial or temporal stability. The findings of the study provide valuable insights for decision-making regarding motorcycle crash prevention strategies at both roadway segments and intersections.

Conclusions: The findings contribute to a deeper understanding of motorcycle crash dynamics and offer valuable guidance for policymakers and traffic safety experts in designing more effective safety measures.