Research On Travel Choice Behavior Of Residents’ Public Transportation Based On A SEM-Logit Model

Urban public transport system with coordinated development of multi-level traffic modes is an effective way to solve urban traffic congestion.Traffic management departments need to understand the influencing factors of travelers’ choice of travel mode and formulate relevant policies to guide people to choose public transportation.There is cooperation and competition in public transport system,so it is very important to understand the selection mechanism and influencing factors of various public transport in the same traffic corridor.For the purpose of studying residents’ public transport travel choice behavior,this paper uses the travel behavior data of residents’ public transport(subway,tram and bus)along Huangpu Road to Zhongshan Road,the main traffic corridor in Dalian,and compares and analyzes MNL model,SEM-MNL model and SEM-ML model,so as to more accurately grasp the behavior law of residents’ public transport travel.Among them,SEM-MNL model considers the impact of psychological latent variables such as operation satisfaction,perceived value and loyalty on Residents’ travel choice.On this basis,SEM-ML model further considers the heterogeneity of residents’ travel.The main conclusions of this paper are as follows:(1)By constructing the SEM model of subway travel,it is found that travelers’ perceived value and life satisfaction will have a positive direct impact on subway loyalty,in which perceived value has the greatest impact on loyalty.(2)By constructing the SEM model of tram travel,it is found that travelers’ perceived value,operation satisfaction and life satisfaction will have a positive direct impact on tram loyalty,among which operation satisfaction has the greatest impact on loyalty.(3)By comparing the goodness of fit of MNL model,SEM-MNL model and SEM-ML model,it is proved that the SEM-ML model,which includes latent variables that are not easy to measure and captures individual heterogeneity,can more effectively analyze the public transport choice behavior of urban residents in the same corridor.(4)Based on the SEM-ML model,this paper analyzes the public transport travel choice behavior of residents in the same corridor.It is found that with the increase of travel cost and travel time,the probability of residents choosing this means of transportation will decrease,and the T test shows that it is reasonable to set the travel time as a random parameter,which also shows that the utility of travel time is different for different individuals.In terms of education,the higher the education level of residents,the more they like to travel by subway.In terms of age,compared with buses,younger passengers are more willing to choose trams.