The Modeling Of Urban Rail Transit Passenger Route Choice In The Condition Of Network Operation

As an important part of public transport,urban rail transit has become one of the effective means to solve the problem of urban traffic,because of its advantages of large transport capacity,high efficiency,punctuality,fast,green and safe.Recently,the urban rail transit is entering the era of network operation,with the accessibility,connectivity and operational flexibility is greatly enhanced,there may be multiple paths for passengers to choose between two stations.The network structure characteristics,for example seamless transfer line network,and the real-time information feedback lead to random travel behavior,path choices of different passengers make the flow distribution more complex and diverse.This paper aims to analysis and model the passenger's route choice behavior under the specified network,which has great significance to the planning and design stage and operation management stage of urban rail transit.Based on the theory of graph and complex network,the paper abstracts the urban rail transit network structure,which can calculate and store the real lines,stations and operation sections,and the relationship between them.The paper also discusses the development process and characteristics of the urban rail transit network,which lays a theoretical foundation for the selection of the effective path and traffic assignment.Secondly,according to the theoretical analysis and questionnaire survey,the paper analysis the factors which influence the different types of passengers when they consider in path selection,besides,the paper analysis the different degree of patience in standing time,transfer time and so on.Then those influence factors are quantified and the generalized cost function is constructed.Based on the characteristics of urban rail transit network and passenger travel behavior,the paper defines and gives the restrictive conditions of the effective path.The paper use the improved Dijkstra algorithm in the searching of shortest path,and discusses the advantages and disadvantages of traditional screening algorithm in effective path,such as the Dial algorithm,K short path,depth first search algorithm.The results shows that Dial algorithm and K short path can not meet the needing of ring network structure,so the paper improves the traditional depth first search algorithm,and puts forward the key nodes,key sections and transfer number restrictions,finally proposes the steps of the modified algorithm.After that,the paper establishes the model of passenger route choice and improves the traditional Logit model.Based on the survey data,the maximum likelihood method is used to estimate the parameters of the Logit model,and the validity and accuracy of the model are tested by using the survey data.Finally,the paper use the urban rail network of Xi'an and a OD matrix of evening peak in weekday to verify the validity of this model,and analyzes the result of distribution.