Exploring contributing factors to crash injury severity at freeway diverge areas using ordered probit model

Abstract

Freeway diverge areas have long been considered crash prone locations. Research is needed to better understand the relationship between the severity consequences of crashes reported at freeway diverge areas and various contributing factors such as geometric layouts and traffic characteristics. The primary objective of this paper is to explore the impacts of contributing factors related to crash injury severity at freeway diverge areas. An ordered probit (OP) model was developed to relate the severity of crash to various explanatory variables using data collected at 326 freeway exit segments in Florida, United States. It is found that the factors significantly influencing crash severity at freeway diverge areas included the mainline lane number, length of ramp, difference of speed limits between mainline and ramp, light condition, weather condition, surrounding land type, alcohol/drug involvement, road surface condition, shoulder width, and crash types of rear-end and sideswipe. The two-lane exit ramps without optional lane were found to have minor crash injury as compared with single-lane exit ramps. The injury severity of crashes do not show significant difference between single-lane exit ramps and single-lane exit ramps with a taper and two-lane exit ramps with an optional lane. Marginal analysis was conducted to quantitatively interpret the marginal effects of contributing factors on each crash severity.