Study On Traffic Flow Distribution Of Stochastic Network With Cumulative Prospect Theory

With the rapid development of economy and growth of vehicles possession, the contradiction between the transportation supply and demand becomes increasingly outstanding. The transportation system is coming into saturation and even super saturation, and the capital and land resources are very limited. Under the uncertain traffic environment, how to distribute and control the traffic flow with limited resource in adequate, reasonable and efficient way becomes an important issue to solve the traffic problems.The behavior mechanism of travelers is considered little in the previous urban traffic studies. The existing urban traffic theories and related models are mostly based on Expected Utility (EU) measurement system with perfectly rational hypothesis, which can’t truly reflect the actual limited rational traffic behavior. To further improve the existing theories, it must fundamentally change the perfect imaginary of travelers. Under the bounded rationality assumption which are closer to objective practice, studying the traffic behavior rules, reviewing urban traffic network equilibrium theory, and establishing more scientific and rational traffic flow distribution models are the key components of this paper.In this paper, multiple theories and technologies, including Cumulative Prospect Theory (CPT), mathematical modeling, Mat lab Programming, parameter sensitivity analysis, have been applied. Considering the limited rational traffic behavior, by introducing travelers’ determination of endogenous reference points based on the budget time, it firstly studies the traffic flow distribution state from the network supply uncertainty and demand uncertainty respectively based on Cumulative Prospect Theory, and then establishes CPT-based stochastic network user equilibrium models. Subsequently, numerical examples are constructed to test the models. Results contrast has verified the CPT is superior to the EU in terms of describing traveler’s decision-making process. And the sensitivity analysis of parameters shows that the effects of networks uncertainty and travelers risk preference on system equilibrium can be effectively expressed through these models. Finally, under the framework of Cumulative Prospect Theory, it discusses the problem of traffic flow control by using tradable credits, which provides a new perspective to the study of managing traffic congestion.