Identifying Precrash Factors for Cars and Trucks on Interstate Highways: Mixed Logit Model Approach

Abstract

This research investigates the factors that lead to three manners of collision (namely, rear-end, angle, and sideswipe) that occurred in the same direction of multilane interstate highways. Mixed logit (MXL) models were developed to estimate the probability of rear-end, angle, and sideswipe collisions as functions of vehicle-following attributes and other driving maneuvers immediately before collisions. The National Automotive Sampling System-General Estimates System crash data set, collected from 2005 to 2008, was used to estimate the model. This research analyzes collisions among passenger cars and trucks, with an emphasis on their vehicular characteristics. Results show that driving behavior is different when vehicular characteristics are different and when roles of the striking and struck vehicles are grouped according to cars and trucks. This research contributes to a better understanding of the differences in unsafe driving acts between cars and trucks, and implications on future policies on car and truck drivers.