Modeling Simple and Combination Effects of Road Geometry and Cross Section Variables on Traffic Accidents

Abstract

This study aimed to find the effects of road geometry and cross section variables on the number of accidents. In addition this study developed a methodology to combine variables using decision trees. Combination variables for road geometry and cross-section variables were developed using the Chi-squared Automatic Interaction Detection (CHAID) algorithm. Three Negative Binomial models were developed: two with homogeneous road segments, and the other with 1-km road segments. The accuracy of Negative Binomial models developed with different road segments was compared. The Negative Binomial model using homogeneous road segments based on horizontal alignment was found to be the most accurate of the three models. Combination variables showed a significant effect on the number of accidents. It was found that the number of accident in a road segment is influenced by the average accident rate in the adjacent road segments.