Predicted the Rate of Extra -Urban Accidents in Ten Iranian Provinces Using Artificial Intelligent based on Influencing Factors

The rising mortality rate associated with intra-urban and extra-urban road accidents constitutes a pressing socio-welfare challenge within the nation. Consequently, this research seeks to forecast accident rates in extra-urban regions across ten provinces in Iran and to pinpoint the primary factors that could facilitate their mitigation. To achieve this objective, Multilayer Perceptron (MLP) neural networks were utilized to assess and determine the most significant contributing factors. Data pertaining to extra-urban accident occurrences, along with essential road enhancement elements, were gathered for the period spanning ۲۰۲۱ to ۲۰۲۳. Upon applying the ANN-MLP model, the findings revealed a prediction accuracy of ۸۵.۳% concerning extra-urban accidents. The MLP analysis identified the total number of operational road maintenance stations, the quantity of technical inspection center lines, and the effectiveness of asphalt distribution as the most pivotal factors affecting extra-urban accidents. The insights derived from this study can aid planners and authorities in recognizing high-risk zones and enhancing road infrastructure. Furthermore, these results lay a robust groundwork for formulating new traffic safety policies and executing preventive strategies, such as improving driver education and fostering a culture of safety within the community.

کلیدواژه ها:

Artificial Intelligence ، Neural Networks ، Classification ، Extra-Urban ، Accident