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 variablesfor road geometry and cross-section variables were developed using the Chi-squared Automatic Interaction Detection (CHAID) algorithm. Two Negative Binomial models were developed: one with Homogeneous road segments, and the other with 1-km road segments. Homogeneous road segments were divided based on the horizontal alignment of the road. Segments were either curved or straight. The accuracy of a Negative Binomial model with homogeneous road segments was compared to that of a Negative Binomial model with fixed-length road segments of 1 km. The Negative Binomial model using homogeneous road segments was found to be the more accurate of the two models. The model with homogeneous road segments was used to draw conclusions on the effect of variables. Combination variables showed a significant effect on the number of accidents. It was found that the road geometry and cross sectionvariables affect the number of accident differently under varying combinations of other variables.
Abstract