Influencing Factors And Decision Mechanism Of Multi-model Traffic Choice Behavior In Urban Agglomeration

The rapid economic and social development has led to a great change in the comprehensive traffic selection behavior in urban agglomerations,which makes it urgent to study the mechanism of multi-mode travel selection behavior in urban agglomerations.On the basis of stochastic utility theory and taking Guanzhong plain urban agglomeration as the research object,this paper determines the factors influencing the travel behavior of urban agglomeration in four kinds of urban agglomeration,such as private cars,high-speed rail,general Railways and high-speed bus,and uses Bayesian estimation method to calibrate the parameters of Logistic model,and establishes the travel selection model of urban agglomeration.Calculate variable OR value,analyze traffic selection behavior of urban agglomeration,and provide basis for traffic planning and operation management of urban agglomeration.Defined the scope and characteristics of urban agglomeration,summarized the whole process of travel in urban agglomeration,combined the development of Guanzhong urban agglomeration and the current situation of traffic,studied from the angles of individual attributes,traveled attributes and distribution attributes,comprehensively considered the influence factors considered by the predecessors in the research,designed the travel investigation scheme,and obtained the relevant data of urban agglomeration travel.Taking the significant degree of variable influence travel behavior as the starting point,this paper made a detailed statistical analysis of the survey data,studied the correlation between each index and the selection behavior,and carried on the multiple collinearity test,and finally determined the model variables.Based on the stochastic utility theory,a multi-Logistic model is used as the model structure to determine the utility function framework,the model is solved by Stata software,and then the model parameters are calibrated by using the traditional maximum likelihood estimation method and Bayesian estimation based on Bayesian theory respectively.According to the parameter calibration result and or value,the behavior of passengers in the multi-mode traffic selection of urban agglomeration is summarized and analyzed.The conclusions are as follows: compared with the traditional calibration method,the model obtained by Bayesian estimation method without priori is more accurate;the control of small car ownership can not effectively reduce private travel;the realization of online ticket purchase can enhance the competitiveness of high-speed bus;the reduction of high-speed rail fares will reduce the passenger flow,but will not reduce private travel;the spatiotemporal accessibility of passenger transport hub has a significant impact on travel behavior.