Commuters' Route Choice And Its Effect On The Urban Transportation System

The urban transportation system contains extreme complexity. Gaining the insight of the system, however, is urgent as the development of urbanization is quite rapid while the supply are limited or is hardly fast increased in a short time. The conflict is more and more serious as time pass. The majority of activities are from commuters in the urban transportation system, obviously. Understanding the rule of commuter'travel activities and the impact on the urban transportation system are critical to implement urban transportation planning and management. The dissertation takes a bottom-up approach and devotes to investigate the effect of commuters'route choice on features of the urban transportation system including spatial distribution of vehicles and the macroscopic fundamental diagram. The main contents are as follows:In the microscopic level, we propose a day-to-day route choice model taking into account commuters'behavioral characteristics First, characteristics of day-to-day route choice and commuters'bounded rationality are analyzed and concluded. Then, a loop architecture based on the evolution of commuters'route choice is presented; the deliberate route choice and habitual route choice in the architecture are discussed in detail. A model is then proposed based on the theoretical analysis. The results of two numerical experiments are presented to demonstrate the performance of the model. The results show that the tendencies of the simulated commuters'choice are similar with real life, and that the model is able to reproduce commuters'route choice with more behavioral characteristics.In the mesoscopic level, we first build a multi-agent macroscopic transportation simulation model on the Swarm platform. In this model, a transportation network is embedded in the lattice space provided by Swarm. Links, intersections and travelers (who steer vehicles) are treated as agents respectively. Travelers are classified by their learning ability. Three scenarios corresponding to various demands, various types of commuters and network structures are conducted. The spatial distribution of vehicles in urban transportation network is discussed using two indices that are degree of unbalance and percentage of links in different magnitude of congestion.In the macroscopic level, we investigate the lastly proposed macroscopic fundamental diagram in day-to-day travel activities. A day-to-day simulation model is first proposed on an open-source traffic simulation platform, A Traffic Simulation (ATS). Two scenarios including different commuters are introduced. The evolution of route choice and the characteristics of the MFD in day-to-day activities, such as convergence, fluctuations and hysteresis phenomena, are analyzed. This investigation provides some basis for controlling the demand to the urban transportation system.