عنوان مقاله [English]
Traffic accidents are one of the main causes of death in the world and the resulting damages of them have significant impact on the economy. In our country either , this issue has become a big and important problem; in the way that In terms of road accidents and the accidents related to traffic our country is introduced as one of the countries with the highest number of accidents and deaths. Various factors such as road conditions, driver characteristics, type of vehicles and their safety equipment and environmental conditions, can affect the probability of occurrence and severity of accidents. Due to the fact that suburban axes account for a large share of accident losses, so in this study, one of the busiest axes in the country - western axis of Khorasan Razavi province - using Friedman and factor statistical analysis and multiple logistic regression models. And the neural network has been examined. The results of this study show that the similar result of Friedman test and factor analysis was this fact that human factor that was identified as the most important effective parameter in road accidents. The results of logistic regression model show that the variables of summer, first days of the week, 12 to 18 hours, night without sufficient lighting, wet and slippy road surface, have increased the probability of accidents in the western axis of Khorasan Razavi province. Also in this model, the variables of 24 to 6 hours, speed less than 50, age over 60 and female drivers have reduced the probability of more severe accidents. The results of the artificial neural network model have shown with higher accuracy that the variables of effective human factor, maximum speed, complete cause of the accident and geometry have had the greatest effect on the severity of road accidents.