چكيده به لاتين
Traffic accidents are caused by a combination of factors that lead to casualties, injuries and property-damages. The four factors of human, vehicle, road and environment always play a key role in the occurrence of such accidents. Among environmental factors, various climatic phenomena also have a significant impact on traffic accidents. The main purpose of this study is to investigate the effect of various factors affecting accidents along with climate factors (including temperature, radiation, rainfall, number of rainy days, wind speed and relative humidity) and important air pollutants (including PM10, PM2.5, SO2, NO2, CO and O3) on the severity of urban accidents in Rasht metropolitan in the years of 1393 to 1397 (SH). 25 variables affecting the severity of accidents have been used along with various methods of analysis and modeling such as Friedman test, exploratory factor analysis, multiple logistic regression model and multilayer perceptron (MLP) artificial neural network model. The results showed that in Friedman test, the variables of weather condition, cause of accident, road surface condition and lighting condition had the highest rank and importance in the occurrence of accidents, respectively. Based on exploratory factor analysis, 25 independent variables were summarized into 7 factors and identified as the main factors influencing accidents and the results of the analysis showed that the variables of weather condition, PM2.5 pollutant, rainfall, road surface condition and lighting condition were under the first effective factor in accidents. Logistic regression model showed that rainy weather and wet pavement surface had the greatest effect on increasing the probability of accidents, respectively, and then PM2.5 pollutant (With concentration of 55.5-150.4 µg/m3) had a significant impact on the occurrence of accidents. The MLP neural network model had better performance and higher prediction percentage than the logistic regression model and its prediction error was lower. The results of the network showed that weather condition, road surface condition, cause of accident and PM2.5 pollutant had the greatest impact on the severity of accidents, respectively, and considering the results of multiple logistic model, rainy weather had the highest influence on urban accidents of Rasht metropolis. Finally, by combining the results, safety solutions were presented to reduce accidents and increase safety in urban roads of Rasht metropolitan.
