Research On Passengers' Booking Behavior Under Conditions Of High Speed Railway Parallel Trains

Recently,to afford a high-frequency and diverse service,the operators of high speed railway implement some measures,such as increasing the frequency and reducing the headway,which emerges the characteristic of mass transit type operation.Typically,when there are several parallel trains that are defined as the trains with different departure times but close running times,passengers can choose the most appropriate one between them as they desired.Through the analysis of the historical passenger booking data,it exhibits the specialty that there are great difference in the passenger booking behavior in the choice of parallel trains.For example,among several parallel trains,some of them are continuous high in booking demands,some of them are in the low booking demands.In this context,the paper are dedicated to researching the passenger booking behavior in the choice of parallel trains to expound the mechanism of passenger booking.The research could provide reference for dynamic pricing and passenger meticulous management of high speed railway system.Aim at the booking behavior in the choice of parallel trains,the main works are as follows:Firstly,the stub records of ticket reservation are used to explain the characteristics of passenger booking behavior with date mining.The train transport capacity utilization can be obtained according to the seat occupancy ratio,and the characteristic of booking behavior of direct passengers with same OD can be analyzed for different parallel trains.Such that the sale horizon can be grouped into several periods and the characteristics of passenger booking behaviors at the particular booking periods can be yielded.Secondly,to figure out the factors affecting the passenger booking behavior,a SP survey was conducted and the questionnaires that include passenger personal attributes,trip attributes and passenger choice preference are designed.The paper focuses on the passenger choice preference and quantifies the preference in the choice of departure time combined with the data with respect to the passenger willingness-to-pay in different departure times,which is of use in basic parameters setting in model establishing.Then,allowing for the preference in the choice of departure time,the generalized cost function which is a utility function of departure time,fare and running time is formulated.Furthermore,the choice model of passenger booking behavior in the case of same fare are established based on maximum utility theory,and an experimental research are conducted to validate the performance of the choice model.Lastly,a simulation application is implemented to simulate the passenger booking behavior in the choice of parallel trains in the case of multi-level dynamic fares.Based on the classification of passengers according to time sensitivity,price sensitivity,and departure time preference,the choice of passenger booking behavior are modelled with respect to different types of passengers,and a dynamic pricing model is then constructed.Futhermore,the process of passenger stochastic arriving and booking via high speed railway are simulated based on the optimal solution of dynamic pricing model,and the simulation results are reflects of the affects of passenger booking behavior since dynamic pricing.