With the rapid growth of car ownership, the scale and the density of the road network are also increasing, but it is followed by large-scale, regional traffic congestion and frequent traffic accidents. Traffic situation risk situational awareness is of great significance to reduce regional traffic operation risks and improve the overall efficiency of traffic operations. Therefore, this paper selects the influencing factors of traffic operation situation in three aspects of road risk, traffic risk and environmental risk, and determines the representative multi-class indicators of the three categories of indicators as the key influencing factors of traffic operation situation. Hierarchical traffic operation situation, using the combination of random forest, FAHP and Fine Kinney to assess the risk of key influencing factors, and establish a traffic operation situation prediction model based on HMM, then verify the accuracy and error of the prediction results. Taking Xi'an Ring Expressway as an example, the forecasting model constructed by the application is used to predict the traffic situation of the road network, and the prediction results are evaluated. The accuracy and error of the three prediction methods of HMM, autoregressive moving average model and gray Markov model are compared. The HMM prediction model proposed in the paper can not only predict the situation value of the road network traffic situation as a whole, but also has higher accuracy and less error.

CICTP