Short-Term Traffic Flow Prediction Based on Bayesian Fusion

With the continuous development of intelligent transportation system, the research about the analysis and processing of road traffic flow have also launched through the prediction of future traffic flow information accurately in real time, the State and government can find appropriate controlling strategy to improve the traffic congestion, in order to make the road network unobstructed and operation efficiently. Therefore, it is of great significance to research the traffic flow prediction of urban traffic system. Based on the full consideration of the nonlinear characteristics of the traffic system, the support vector machine model and the BP neural network model optimized by genetic algorithm are used to analyze respectively the traffic flow of the road segment. Within this framework, the Bayes fusion is proposed to address the limitations of a single method used for prediction. The performances of proposed methods are evaluated by an experimental application with the measured data on the main road in Tangshan. The results show that the proposed prediction method solves the limitation of single method prediction and has higher prediction accuracy.