Tangent length and sight distance effects on accident rates at horizontal curves on two-lane rural highways

Abstract

This thesis documents an evaluation of the relationships between accident rates at horizontal curves and preceding tangent length and sight distance. Data collection and statistical methods used to evaluate this relationship are presented. A base relationship between accident rates at horizontal curves and degree of curvature was established, and the effects of various variables on this relationship were examined. These variables included lane width, total pavement width, state in which the data was collected, tangent length and sight distance. Analysis was performed to determine whether these variables should be included in an operating-speed or workload based consistency model. The analysis of lane width, total pavement width, and state demonstrated no practical significance for explaining the variability in the relationship between accident rate at horizontal curves and degree of curvature. Therefore, none of these variables were included in the final model. This thesis included two hypothesis: (11) tangent length is a significant factor that should be included in an operating- speed or workload based consistency model, and (2) preceding sight distance is a significant factor that should be included in an operating-speed or workload based consistency model. Regression analysis on this database suggested that neither tangent length nor preceding sight distance are significant variables in explaining the variability in accidents on horizontal curves for a given degree of curvature category. Results, however, did verify that degree of curvature is a strong predictor of accident rates on horizontal curves.