Study On Commuting Choice In Suburban Multi-modal Traffic Network

With China's sustained and rapid economic development, accelerating the process of urbanization, large numbers of people quickly gathered, causing traffic problems in mega-cities. In order to solve the traffic problems in the peak period, subway, light rail, BRT in the major cities are rapidly built or expansion, the original form of traffic methods constantly improved, multi-model traffic system gradually improved. How cost-effective use of their research is the weak link? Commuter travel direct leads to the city's early and evening peak, the paper set on the eyesights of users of the transportation, research the urban residents travel choice model, algorithm and the practical application.This papers first reviewed the development of the non-collection model theory, summed up the modeling basic ideas, classical model and model development systematically. Based on this, this paper choose Nest Logit model which suitbility use in research the multi-modal transport network in the suburban area.This paper further do some research on studying commuter transport network, with the idea of graph theory differienciate travel chain in multi-modal transportation network into direct link and transfer link then use the state variables to mark the differient state of the node. At the same time, analysis the characteristics of commuter traffic, construction network of commuter traffic, and use Nest Logit model in multi-modal commuter transport network.In order to calibration the above model, the paper proposed Newton's continuation algorithm for maximum likelihood function approach. Through the investigation of Shanghai new suburban city, Zhangjiang Town commuter travel choices, data collection, based on the real example, designed a calculation process using Matlab. Discuss the steps of the model using in the practical application and the problems that needs to pay attention.Final this paper extension of the application of the model. Based on the contribution rate forecast, according to the change of the cost, quantitative evaluation the changes of the traffic service attribute cause by the quality of traffic sub-network service change. At the same time, summed up the changes of income of the traffic operations department. This paper do some beneficial attempt on big citys transportation network optimization, travel choice model set up and its key technology realization.