Study On Characteristics Analysis And Modeling Of Intermodal Travel Of Regional Integrated Transportation

The research of intermodal travel is instrumental in developing the comparative advantages of different modes of transport and improve the operation efficiency of the regional integrated transportation portfolio.This article relies on the integrated traffic analysis technology of the comprehensive transportation system,which is a key project of the National Natural Science Foundation of China,and deeply explores the characteristics of intermodal travel in the regional integrated transportation system as well as mode choice modeling.We firstly analyze travel patterns of some modes of transport,laying the groundwork for the design of subsequent questionnaire scenarios and attribute levels.According to the advantage of different modes of transport,we design short,medium and long-distance intermodal travel scenarios by a uniform method.A total of 3163 questionnaires are collected after excluding invalid ones.We make a preliminary statistical analysis on the characteristics of the intermodal travel passengers.Based on the SP surveys,three discrete choice models,Multinomial Logit(MNL)、Nested Logit(NL)and Mixed Multinomial Logit(MMNL),are used to study the mode choice behavior of the passengers under the three scenarios with different distances.We perform the quantitative analysis of the effects of individual attributes,alternative-specific attributes,intermodal travel habits and attitudes(including travel tolerance,satisfactions of modes of transport and concern of intermodal service)on the mode choice behavior.Because of considering random effects and individual heterogeneities of the key variables,the MMNL is the best compared with NL and MNL,and has a strong explanatory power.The model with the best performance is selected to analyze the internal law and mechanism of mode choice in intermodal travel.The results show that all four alternative-specific variables,i.e.,intracity travel time,intercity travel time,total cost and transfer time,have significant negative influences on the utilities.Intermodal passengers pay more attention on the intracity and transfer time under the short-distance scene.The intercity travel time will have a larger impact on travelers with the increase of the travel distances.They are more sensitive to the total cost under the short and medium distance scene compared with the long scene.The other individual attribute variables,intermodal habit variables and attitude variables have different degrees of impact on the mode choice under the three distance intermodal travel scenarios.Furthermore,based on the different preferences of travelers,we put forward some intermodal travel development suggestions in terms of hardware facilities,service quality,and encouragement policies.Finally,on the basis of comprehensive transportation version of Tran Star,we use the MMNL model as the kernel to design the module of mode prediction of intermodal travel.This module consists of two functions,simple prediction on engineering practice and high-precision prediction,by considering the demand of different users,which lays the foundation for implementing the projects of predicting the modes of transport in intermodal travel.The research of intermodal travel is instrumental in developing the comparative advantages of the.