The Research Of Transit Assignment Modeling And Its Application

With the rise of the planning theory about transport sustainable development, the role of public transport system in urban transport system has received more and more concern all over the world. Transit assignment, as the core of the transit planning and analysis process, has been developed at a dramatic pace.At the same time, the fact that simulation technologies have been introduced into transport planning field offer transport planner better tools to check problems, work out solutions and forecast the future trends. However, the unique characters of transit assignment, vast and complex transport modeling theory system, along with the distinctions between models, give rise to a brand new problem that modelers ought to confront with; that is how can one select and build the optimal model that could guarantee the correctness of model use and the closest similarity between the model result and the reality. On this background, this paper has focused on the transit assignment theories, and tried to set up the structure of model choice.Besides detailed introduction and analysis of different transit assignment theories, functions comparison between EMME/2 and VISUM is also a focus of this paper. In order to clarify the characters of different models, bus network and a simplified network was built for the research on the function of capacity constraint in transit assignments, the comparison of two different kind of schedule-based assignments, and the comparison between frequency-based and schedule-based transit assignments. In the end, combined the advanced overseas experience and our model comparison results, this paper presented the influence factors of model and software choices with the aim of guide and help modeler better understand model characters, and run transit assignment models successfully.This paper is the result of thorough research on the transit assignment modeling and efficient solutions to transit assignment and related areas. Our study should also be proved helpful, applied into other modeling research and model choice decision-making process.