Study Of Travel Mode Choice Behavior Based On Disaggregate Model

In recent years, with the development of urbanization, travel difficulties become increasingly prominent. Transportation structure of the city is formed by individual travel mode choice. Transportation mode selection is an important part in transportation planning and policy-making. Currently, under the level of socio-economic development of the city, we need study residents of different modes of probable transportation utilization. It could predict the mode of transportation objectively and effectively. It can provide the scientific basis for decision making and urban transport structure optimization. In recent years, Chinese cities are beginning to realize that it is important to master the law of the residents travel. It has a very important role in urban transport planning, organization, management and control. It is also appropriate to carry out the residents travel survey in order to study the influencing factors.The paper is based on the actual situation of urban traffic surveys and relevant planning information collation. It is investigated on the regular bus, BRT, taxi and subway and private transport, while the main characteristics of urban transportation modes and influencing factors are described.For current travel choice models considered less latent variables and the effect of latent variables on the selection result, the paper portrays latent variables through structural equation modeling. Then the paper combines maximum utility theory, it improves the utility function of the traditional Logit model, and constructs SEM-Logit integration model with latent variables. Use SPSS, Smart PLS and other professional software to solve the model integratedly. After determining the choice model, the article distinguishes the main factors affecting the choice of transportation.According to the model constructed, collect questionnaires of the residents transportation options in Nanjing city. First, statistic the questionnaire datas, and questionnaire datas of latent variables are processed by Smart PLS. It could obtain latent variable scores. Finally use SPSS to calculate logit model of all variables. It has the comparative analysis of main factors of residents travel. On the basis of that traffic travel-related variables are analyzed, and it finally gives the relevant measures to improve public transport.