Risk Assessment of Traffic Facility on Freeway Based on a Modified Bayesian Network Model

Traffic facilities play an important role in traffic safety.Assessing the risk of facilities on freeways is one of the effective methods to prevent traffic accidents.In order to evaluate the risk, this study analyzes the effects of facilities on freeway safety.The nodes and network are extracted by factor analysis.According to the relationship of nodes in a network, combined with probability theory (such as the conditional probability), this paper modifies the Bayesian formula after rounding off unrelated variables and no-effect constants, and establishes a modified Bayesian model for assessing facilities on freeways.With the traffic data of freeways and the result of a survey, the risk of facilities can be assessed.The results computed by the modified model show that the smaller the value, the higher the risk of freeway facilities.Using the data of Huiwu freeway to verify this proposed model, it shows that when the value is smaller than 0.5, the number of crashes is relatively larger, correspondingly, the risk of facilities is higher;otherwise, the risk of facilities is lower.