Research On Road Traffic Safety Index Based On Bayesian Network

In recent years,with the significant improvement in motorization,road traffic accidents have also occurred frequently.The number of casualties caused by traffic accidents has remained high for a long time in the past.Therefore,analyzing the influencing factors of road traffic safety through scientific and effective methods,in order to evaluate the road traffic safety level,and put forward corresponding improvement measures,is essential to ensure road traffic safety.The thesis first describes the structure of road traffic system and the characteristics of traffic accidents;the influencing factors of road traffic safety are analyzed from five aspects of people,vehicles,roads,environment and management;the construction principles of the road traffic safety index system are proposed and the road traffic safety is established Evaluation System.Secondly,based on the index theory,the concept of road traffic safety index is proposed,and the road traffic safety level and the threshold of level index are determined.Thirdly,a Bayesian network-based road traffic safety index model is constructed.This model uses the explanatory structure model method to determine the relationship between influencing factors,and uses fuzzy set theory to convert natural language variables into precise probabilities.Finally,taking the road traffic in Wuhan as an example,the road traffic safety level in Wuhan was evaluated,and the rationality and accuracy of the road traffic safety index to evaluate the road traffic safety level were verified.The research in this thesis shows that the Bayesian network-based road traffic safety index model can reflect the road traffic safety status and can diagnose the deficiencies in the traffic system,providing a way of thinking and methods for the traffic management department,which has certain Practical application significance.