Research on Passenger Flow Assignment Method Based on Complex Train Service Network

The attritutes and constructing method of the complex train service network(CTSN) formed on the basis of passenger train plans were studied.By analyzing multi-level passengers' choice behavior,the arc resistance of CTSN was determined.The passenger flow assignment model of CTSN reflecting diversified passengers' travel demand was estabished.The model was made to take the maximized passenger travel utility as the objective.The multi-constraint conditions of passenger travel routes were set by respectively restricting the number of transfer times and the transfer time of passengers with different travel distances and different levels of travel demand.The hybrid algorithm consisting of the improved ant colony and Frank-Wolfe algorithms was designed to solve the model.In order to match the choice of rail passenger behavior,strengthening of passenger flow was assigned by rules of OD traffic,trip distance and priority.Finally,case study on passenger train plans of part of the network centered by the Beijing-Shanghai High-speed Railway was made to accomplish passenger traffic assignments and the effectiveness of the proposed model and algorithm was proved.