Implementing a New Intelligent Adaptive Nero Fuzzy Inference System (ANFIS) based on Identifying and Predicting Road Accidents
Malek Mohammadi, Fatemeh Alsadat Zaeimbashi, Morteza Rahmani Nikouei
Nowadays, the number of road accidents in our country is increasing and since the financial, life, psychological and social damage in some cases are irreparable it is necessary to prevent traffic accidents damages the effective factors identify and effective solutions to be implemented. The main purpose of this research, is study of various (natural and human) causes of road accidents on the road of Haraz and providing a method for identifying incidental factors based on(FMEA) method and eventually predict events Risk Priority Number (RPN) based on the Adaptive Neural Fuzzy Inference System (ANFIS) using MATLAB software. The results of the failure analysis method (failure modes & effects analysis) (FMEA) show that the two factors of mucus and the consumption of narcotics and alcoholic beverages are the most effective factors in the accidents of Haraz road and also due to the negligibility of predictive error values, Mean Squared Error (MSE) and Root Mean Square Error (RMSE) the accuracy of the prediction by the adaptive neural -fuzzy inference system is confirmed. The result of this prediction system (ANFIS) can be effective to predict the number of road accidents risk and providing solutions to reduce the road accidents.