Study On System Optimization Based On Travel Mode Choice And Traffic Policies

With the development of society and economy, vehicle population is growing, which makes the traffic contradiction between supply and demand more prominent. Facing with a variety of travel modes, the fundamental measure of relieving traffic congestion is to control the utilization of private cars by effective traffic demand management strategies and attract more transit users by improving public transport services level. Within the framework of cumulative prospect theory of behavior economics, this paper intended to analyze mode choice behaviors under uncertainty and study the congestion pricing, pricing refund and rationing policies. Specifically, the main contents of this dissertation are summarized as follows:(1) Mode choice behavior based on cumulative prospect theory was studied. This paper applied cumulative prospect theory to explore mode choice behavior. Regarding the expected cost of one travel mode as the reference point of the other travel mode, The travel utility in the Logit model was replaced by the travel prospect value. Then, the mode choice model based on cumulative prospect theory was developed. Theoretically, the influences of model parameters, traffic demand characteristics, transit fare and congestion pricing on the mode choice and users’total travel cost were illustrated.(2) Considering travelers’mode choice behavior, the congestion pricing and refund policy is studied. On the foundation of cumulative prospect theory-based mode choice model, bi-level programming model was proposed for seeking for optimal congestion pricing and refund scheme and solved by step acceleration method and penalty function method. A numerical example was presented to illustrate the influences of different transit fare discount rate and congestion pricing level on the system travel cost and expense cost. The numerical example was also employed to compare the differences of system travel cost/travel modal split rate among three situations:subsidy and no congestion pricing, congestion pricing and no subsidy, congestion pricing and subsidy to no congestion pricing and no subsidy. It was found that the congestion pricing and revenue redistribution policy can not only reduce the total system travel cost but also save the expense cost of transit users and the fiscal expenditure of government.(3) Considering travelers’ mode choice behavior, the traffic rationing strategy is studied. With the difference of travelers’ value of time, the problem of travelers’ mode choice between private car and transit before and after the implementation of rationing strategy was studied. The bi-level programming was proposed, of which the upper-level objective was to minimize the total system travel cost, as well as the lower-level problem was the Multinomial Logit travel mode choice model. The model was solved by particle swarm algorithm and the optimal departure frequency and rationing ratio before and after the implementation of rationing were obtained. A numerical example was presented to verify the Pareto Optimality and Economies of Scale of transit and analyze the impact of rationing policy on the modal split and the total system travel cost.