Application of Combined Traffic Demand Forecasting for Comprehensive Transport Corridor

Traffic demand forecasting is the indispensable process of capacity and resource optimal allocation of comprehensive transport corridor. As such, it is derivate by social and economy activities. The steps of traffic demand forecast of comprehensive transport corridor in this paper is as follows; Step1: Using principal component analysis as a pretreatment to filter out the critical factors that affect traffic demand in corridor; Step2: Taking correlation coefficient method to forecast traffic demand in corridor; Step3: Forecasting based on BP Neural network; Step4: Adopting variance-covariance method to integrate above two forecasting results, then the result is the weighted average of the two methods. It is proved that not only vertical trend demand based on time sequence but also the horizontal demand based on social economic factors are taken into account in this model. Moreover, the deficiency of each method is overcomed and the advantages of these methods are integrated at the same time, which could improve the accuracy of forecasting and set up a solid basis for next capacity optimization.