Analysis of Factors Affecting Driving Violations Based on Artificial Neural Network

The purpose of this paper is to present a model based on neural networks with high accuracy to build a model for predicting violations based on factors such as a vehicle, human factors, environmental factors. Considering that in most studies, the role of the road and the vehicle have been considered more than in other cases, in this research, a model has been designed using various factors. The approach of this research is combined (quantitative, qualitative) and in terms of applied purpose and descriptive survey method, using multilayer model neural network (MLP) has been used. The effect of the transition on traffic violations has been studied so that the optimal model is selected and the three-layer model with high accuracy of 0.90483 is predicted as a model. The findings of this study indicate that human factors including gender, age, education, Occupation has the highest share of traffic violations and the three-layer model has better results and is more consistent with real data.