Research On Passenger Flow Assignment Of Urban Rail Transit Based On Passengers' Familiarity With Rail Transit Network

With the development trend of urban rail transit network operation,the rail transit network scale of each city is continuously expanding and the rail transit network structure is getting better and better,more and more paths are available for passengers between terminal stations,but at the same time,the diversity of passenger path selection also increases the uncertainty of the passenger flow distribution in the rail transit network,making the operation and management of urban rail transit face great challenges.Based on this,it is essential to model passengers' travel behavior theoretically,calculate and analyze passengers' path selection result and passenger flow distribution of rail transit network,providing a strong theoretical support for the optimization of urban rail transit operation and organization,and the ticket clearing of the rail transit system.Travel behavior of rail transit passengers is closely related to passenger attributes,and the familiarity degree with rail transit network is the key attribute affecting passengers' path selection.Therefore,based on passengers' familiarity degree with the rail transit network,this paper focuses on the passenger flow assignment of urban rail transit,and carries out the following research work:(1)To master the path selection behavior characteristics of passengers with different rail transit network familiarity degrees,the passenger travel behavior questionnaire is designed by the method combining the RP survey with the SP survey,and the travel information and path selection willingness of passengers are investigated.The survey result shows that passengers are more sensitive to the travel time than other factors,passengers unfamiliar with the rail transit network are significantly more sensitive to the transfer times than passengers familiar with the rail transit network.(2)Based on passengers' familiarity degree with the rail transit network and the path selection basis,passengers are classified as passengers familiar with the rail transit network,class I passengers unfamiliar with the rail transit network(passengers unfamiliar with the rail transit network traveling by the subway line diagram)and class? passengers unfamiliar with the rail transit network(passengers unfamiliar with the rail transit network traveling by the map APP software and inquiry),the generalized cost function of three types of passengers are constructed separately.For the passengers familiar with the rail transit network and the class ? passengers unfamiliar with the rail transit network,the generalized cost function is constructed according to the travel time,transfer times,transfer time and the degree of congestion,but the parameters need to be estimated separately.For the class I passengers unfamiliar with the rail transit network,the generalized cost function is constructed according to the travel information provided on the subway line schematic map,including the length of the path,the number of stations and transfer stations on the path.(3)The urban rail transit path selection model based on Multinational Logit Model is established,and the parameters of the model are estimated by multiple linear regression method combined with the survey data.The passenger flow assignment algorithm based on the familiarity of passengers with the rail transit network is designed,and three validation methods of model algorithm are proposed.In this paper,the questionnaire survey test method and cross-section passenger flow test method are used to verify the results of passenger flow assignment.(4)Taking Xi'an Urban Rail Transit as an example,the rail transit network topology diagram is constructed and the basic data required for passenger flow assignment are organized,based on the passenger flow assignment model,the passenger flow assignment at evening peak period of Xi'an Urban Rail Transit is completed.Aiming at 4284 OD pairs in the rail transit network,5163 effective paths are searched out,and the results of effective path selection among OD pairs in rail transit network are obtained.Two OD pairs of intention survey pairs are selected to analyze the probability of passenger's path selection with different familiarity degree of rail transit network.It is found that passengers unfamiliar with the rail transit network prefer to travel on the path with fewer transfer times and shorter transfer time than those familiar with the rail transit network,and their sensitivity to timeliness is relatively low.By superimposing the flow of different types of passengers,the cross-section passenger flow and the transfer passenger flow are obtained by statistics.It is found that the average section flow of Line 2 is the largest and the unbalanced degree of section discharge and direction of Line 3 is the largest.The transfer pressure of Beidajie Station is the largest,Xiaozhai Station is the second,and Tonghuamen Station is relatively small.By comparing the calculated resuit of path selection with the data of travel intention survey,and comparing the result of section passenger flow assignment with the actual section passenger flow data,the validity of the passenger flow assignment model and algorithm based on the passenger's familiarity with rail transit network is verified.