Study On Activity-based Travel Demand Forecasting Model

As we all know, urban transportation system is a big complicated system. Thepractice has proved that, in order to resolve the traffic congestion problem,increasing financing and accelerating transportation infrastructure construction isnot the ultimate method. Especially in our large urban cities, the framework ofroad building has been basically ensured and the land resources that can be usedhave already been limited. With the development of economy and accelerating ofurbanization, the residents in rural areas are swarming into the cities, people'stravel demands are increasing quickly. Although the transportation infrastructure isbeing increasingly financed, the traffic congestion problems are becoming moreand more serious. In recent years, transportation planning and constructingdepartments began to realize the contradiction and intend to resolve it. On onehand traffic supply should be increased properly, on the other hand travel demandsshould be restrained and adjusted so that the development of the transportationsystem can be balanced. This is an efficient method to settle traffic congestionproblem. Consequently people attach more and more importance to the theory ofTransportation Demand Management (TDM). TDM aims to control the total quantity of traffic demands so as to gain theappropriate transportation construction scale. That is, we can achieve the balanceof traffic supply and demand and relieve the traffic congestion by directing andadjusting the time-space distribution of traffic demands so that the transportationsystem can work efficiently, the passengers and cargos can reach their destinationsquickly and safely, and at the same time urban environment can be improved aswell. The core of TDM is to settle traffic congest problem by directing the wayshow people travel. But the practice has proved that not all the TDM policies canachieve the anticipated effects, some of them are not accepted by the travelers ormislead them. Therefore researchers begin to study residents' travel characteristicsand analyze their travel behaviors by travel demand forecasting so that the traveldemands can be forecasted accurately.Based on these backgrounds and combined with the Specialized Research Fundfor the Doctoral Program of Higher Education (SRFDP), the activity-baseddisaggregated travel choice model (20030183008) and the National NaturalScience Foundation of China (NSFC), the activity-based travel behavior analysismodel and TDM policy simulation evaluation method (50578094), this paperattempts to analyze the traffic condition, the social and economic environment,and residents' standard of living in our country firstly. Secondly it focuses ondeveloping the activity-based daily activity pattern model, time choice model andmode choice model. Finally it aims to put them into studying and evaluating someTDM policies. With these researches, it tries to explore a kind of travel demandforecasting method conforming to the development of transportation system in ourcountry, and provides a new method for transportation planning and TDM policyevaluation. This is the purpose of this paper. The main contents of this paper canbe summarized as follows:1. By analyzing the development of travel demand forecasting models, thesorts of the models and the activity-based modeling ideas in our country and theworld as well, this paper summarizes the basic theories and modeling methods ofthe activity-based travel demand forecasting model, and introduces the applicationof this model to the validity evaluation of the transportation demand management.With the analysis and summary, it establishes the framework of the activity-basedtravel demand forecasting model system.2. This paper summarizes the basic sorts of the activity-based forecastingmodel and introduces modeling method and main characteristics of each one. Itdescribes the basic idea and models-MNL, NL and Logistic of the disaggregatedmodel, and introduces the configuration and estimation method of each one indetails.3. According to the modeling demand of the activity-based travel demandforecasting model, this paper analyzes the distribution characteristics of residents'traveling in the respects of time, space, mode, age and so on. With the analysis, itsummarizes the main reasons of the traffic congestion in Changchun city.4. By analyzing the traffic and economic developing condition in Changchuncity, it develops activity-based daily activity pattern model, travel time choicemodel and mode choice model, estimates each model with statistic software SPSS,and finishes testing and analysis of each variable using sensitivity analysismethod.5. It introduces the concept and the main measures of the transportationdemand management, studies the application of the submodels of theactivity-based travel demand forecasting model system, compares and analyzes thechanges of residents' travel characteristics before and after implementing the twoTDM policies of congestion pricing and the public traffic policies, with which toevaluate the feasibility of the policies and to provide basis for making andimplementing similar TDM policies.Combined with the study objectives and contents of the two fund projects,this paper aims to develop activity-based travel demand forecasting model system,and studies primarily how to evaluate and analyze the anticipated effects of theTDM policy using the models. The main innovations of this paper are:1. It develops an activity-based travel demand forecasting model systemcomposed of a daily activity pattern model, a travel time choice model and a modechoice model.2. By summarizing domestic and international activity-based modelingexperiences it develops a new NL model formed by several Logistic models, andfinishes the estimation, variables validation and connection of the three models ofthe system using the MNL module in the SPSS statistic software.3. By using the daily activity pattern model, travel time and mode choicemodel, it predicts and compares the daily activity pattern, travel time and modeadjustments made by the residents because of implementation the two TDMpolicies of congestion pricing and the public traffic policies. With the analysis, itevaluates the feasibility of the policies, and provides the basis for establishingsimilar TDM policies.