Identification and Prioritization of Effective Parameters in Injury-Causing Accidents at Four- Lane Intersections (Case study: Isfahan City)

Injuries and damages caused by traffic accidents in road networks have become a dilemma. According to statistics, a significant part of the total number of accidents occurs due to the convergence of traffic flows at intersections. The aim of this study is to investigate the factors affecting the safety of the intersections, evaluate them, and provide a statistical model for predicting the number of accidents and ultimately provide solutions for the safety of intersections in order to reduce the number of injured people at the intersections.

In order to predict the number of traffic accidents based on traffic parameters, geometric design and traffic control features, three statistical models were used in this study: Poisson Log- Linear, Negative Binomial Log-Linear, and Linear Models. Modeling and analyzing data in this study was done based on SPSS software and Excel. The goodness of fitting was controlled by Deviance/Degree of Freedom and x2 Pearson/Degree of Freedom.

According to the results, the reduction in volume, exclusive right turning lane, camera and increase in length of the exclusive right turn lane, reduce crashes by 190.37%, 52.06%, 27.09%, and 1.06%, respectively. As the mid-block width in the intersection increases by 1 m, the frequency of accidents will increase by 2.02%. Adding a lane in an intersection will lead to an increase in accidents by 18.55%. If two intersections have completely identical conditions and only the number of phases differs, then an intersection with two phases will have a reduction of crashes of 35.649% over the intersection of three phases. Therefore, it is necessary to take appropriate steps regarding to strength of each variable in the reduction of crashes.

According to the results of the Poisson model, it has a better accuracy than the negative binomial model. It was also found that the seven above-mentioned variables affect the safety of intersections, which is suggested to be considered in urban immunization programs.