Research On The Public Transportation Trip Mode Choice Behavior Based On Random Parameter Logit Model

Vigorously developing public transport and improving public transport travel sharing rate is the inevitable choice to solve the problem of urban traffic congestion in large and medium-sized cities in China.In order to effectively improve the public transport sharing rate,it is necessary to study the travelers' trip mode choice behavior.In the past,there are two main shortcomings in the study of travelers' travel mode choice behavior: one is that most of them only pay attention to the characteristics of traffic mode,such as fare,time and other factors,while ignoring the latent variables,the factors that can not be directly observed or measured,these latent variables also have an impact on travelers' trip mode choice behavior;the other is the traditional discrete choice models established before by the research on choice behavior mostly represent individual preferences with group average preferences,ignoring the differences of individual preferences.Therefore,in the study of travelers' choice behavior between metro and conventional public transport modes,this paper adds six latent variables of travelers' demand and subjective evaluation for vehicle characteristics,including travelers' requirement of vehicles' comfort and reliability,travelers' evaluation on metro and conventional public transport' comfort and reliability,and based on the random parameter logit mode to study ravelers' trip mode choice behavior.The main contents of this paper are as follows:Firstly,by analyzing the whole process of trip choice behavior,analyzing the factors that influence travelers' trip choice behavior.And the related theories of Logit model are summarized.Secondly,collecting data through questionnaire survey.First,representing latent variables which can not be observed or measured directly by index variables and setting corresponding measurement questions in the questionnaire.Then,designing a questionnaire combining SP and RP.Finally,analyzing the collected questionnaire data.Thirdly,expounding the theory of structural equation model,and constructing the structural equation model which reflects the relationship between latent variables and indicators,and calibrate latent variables based on the constructed structural equation model.Fourthly,based on the random parameter logit mode to study ravelers' trip mode choice behavior between conventional public and metro,adding this six latent variables related to traveler's subjective evaluation and requirements for vehicle characteristics into the random parameter logit model,to study the effects of fare,in-car time,travelers' evaluation of vehicle comfort and traveler's evaluation of vehicle reliability on traveler's trip mode choice behavior.The results show that fare and in-vehicle time have a negative impact on utility,while travelers' evaluation of comfort and reliability of vehicles has a positive impact on utility.The older the travelers are,the smaller the impact of fare and in-vehicle time on travelers' utility;The higher the education level of travelers,the smaller the impact of in-vehicle time on travelers' utility.The higher the traveler's requirement on the comfort of the vehicle,the greater the impact of fare and in-car time on the traveler's utility.