Transit Passengers’ Travel Routes Identification And Choice Behavior Analysis

Developing public transit and facilitating the travel of urban residents are key measures to alleviate traffic congestion in most cities in China.When a public transit system is in a good condition,passengers can complete their trips safely,conveniently,accurately,efficiently and economically on the transit network.With the development of modern society,economy and technology,passengers have undergone many changes in their travel psychology,travel mode and choice behavior.Therefore,the trasitional shortest travel path of public transit is often difficult to describe and guide people’s actual travel problems.First of all,the data of this paper is composed of public transit big data(passengers’ IC card,GPS data of buses,public transit routes and station information).On the basis of data fusion,station identification algorithm is used to infer passengers’ boarding stations and alighting stations.This paper identifies the passenger flow distribution of boarding and alighting at bus stops in Xi’an during working days.Then design algorithms to identify transfers based on passengers’ characteristics,and match the overall passenger OD distributionSecondly,according to the definition of public transit travel chain and travel path identification based on station recognition,passengers’ trajectory and bus OD distribution are extracte.Then construct the actual travel route selection set in the public transportation network,and use the location matching algorithm to generate the actual travel route of the passengers.While matching the actual travel path(ATP)with the map,the theoretical shortest time travel path(STTP)and shortest distance travel path(SDTP)are also calculated.The latter uses the travel time and the length of the road section as the impedance function to construct the shortest distance path and the shortest time path model.Thirdly,compare these three paths by calculating their overlap distribution,the difference in travel time and distance,the distribution difference between the ATP and SDTP and STTP in various travel attributes.A multiple linear regression model is built for checking the differences between these paths.The path deviation influence analysis focus on the distance ratio,the time ratio and the degree of overlap.Combined with the existing analysis and research on the characteristics of bus passengers,this article attempts to analyze why travelers not choosing the shortest route from the aspects of selflessness and rationality.Finally,this paper proposes a bus passenger route selection model based on the Path Size Logit.Using the Xi’an bus passenger data collected in this paper,the parameters of the choice behavior in the route selection model are calibrated.And the influences that related to the choice of passengers’ routes and their use of non-shortest routes are also found.This article starts from the non-theoretical shortest travel path of bus passengers,which is different from the previous analysis and research of passenger characteristics of a single bus line or a single bus stop.The use of big data-based travel chain restoration provides a way of thinking for studying the reasons for actual non-shortest travel path.