Traffic Forecasting based on the Combined Corecasting Model

Minxin Lu*, Jingyang Sun

School of Economics and Finance, Anhui University of Finance and Economics, Bengbu, 233030, China

Abstract: In this paper, aiming at the problem of traffic jam time prediction, in order to determine the more accurate traffic jam time under the road, four indicators of road level, traffic flow, time period and road length in Internet road condition data are selected as input variables, so as to establish BP neural network and SVM model to predict the travel time of the output variables. Considering the disadvantages of using only a single prediction method, the multi-model fusion prediction algorithm (MMFA) was selected to improve the prediction accuracy of the output variables by using the output information of multiple prediction methods. The BP neural network model and SVM model were combined to obtain a relatively accurate prediction value of the travel time.

Keywords: Clogging time prediction; Traffic flow; BP neural network; SVM model