Study Based On The Bayesian Network For Highway Traffic Emergency Situation Assessment

With the increase of highway traffic mileage, the transportation industry in our countryhas entered the rise stage of emergencies of and the social impact of traffic incidents isgrowing. In order to deal with highway traffic incidents, traffic authorities developedcontingency plans, but due to the lack of effective assessment of the traffic incidents, a lot ofplans have the accuracy and operability problems. Therefore, how to assess the traffic incidentsituation effectively and develop a reasonable emergency policy is the key to deal withemergencies.After analyzing achieve technology of situation assessment, based on comparison of theadvantages and disadvantages of various methods to determine the hierarchical Bayesiannetwork as highway traffic incidents situation assessment method. Building bayesian networkmodel which start from the highway traffic emergencies type, causal factors, characteristicsand evolution mechanism, combined with the characteristics of Bayesian networks, and useD-S evidence theory and the joint tree,andbeing the instances of authentication by transportaccidents in production.The main research work and conclusions:(1) By analysis of the characteristics of thehighway traffic emergencies, such as the types,disaster-causing factors and evolutionmechanism, combined with the characteristics of the situation assessment, study theapplicability of situation assessment inthe traffic emergencies. The results show that thesituation assessment concept can be applied to highway traffic incidents which has a strongguiding role.(2) Using bayesian networks to build a highway traffic emergency situationassessment model. The model is feasible in actual emergency.