Modeling the Prediction of Fatal Crashes at the Signalized Intersection (Case Study: Isfahan)

Intersections are a factor in determining the capacity of the transportation network, and any disruptions such as accidents in them will result in a severe capacity reduction. Accident crashes at intersections are significant due to the convergence of traffic flows. The purpose of this study is to investigate the factors affecting fatal accidents. For this purpose, geometric, traffic, fatality and control status data of 65 intersections of Isfahan city were extracted and analyzed in SPSS22 software. This analysis was carried out using Poisson linear logarithmic modeling, negative binomial linear logarithmic and linear regression. The results showed the superiority of the negative binomial model to the other two models, and it was found that 5 variables including the number of phases at the intersection, the presence of the camera, the width of the line, the width of the left-turning lane and the straight transit volume, are among the variables that affect safety.