Accidents modeling and pedestrian safety on suburban roads using logistics model

Pedestrians are the most vulnerable part of the traffic in suburban streets, and according to the statistics of pedestrian accidents, identifying the factors affecting accidents in suburban roads which can lead to decreased incidence and severity of accidents is clearly understandable. In this study, the detailed implementation of the 1023 forms of pedestrian accident information was performed in Rasht – Lahijan road, and using the logistics model, the probability of occurrence of the accident was calculated for each category. For this purpose, different methods of modeling such as backward, forward, and entering were used and evaluated. Finally, the model of the backward method with R= 0.576 was chosen as an optimal model. The results showed that according to the proposed model for pedestrian accidents in Rasht - Lahijan road, parameters such as time of accident (12 am – 24 pm), rainy weather condition, final cause (due forward), face to face conflict, and guilty and non-guilty vehicle (KIA - Pride) of accidents led to injury or death.