连续型异质有限理性用户分类及其随机网络均衡

为提高交通行为引导政策的针对性,有必要研究随机路网环境下不同类型出行者对交通引导政策的反应,即多类有限理性用户风险态度和路网退化系数对用户行为决策的影响.本文构建了连续风险态度离散化的用户分类模型以及基于退化路网的多类有限理性用户均衡模型,证明了解的存在性,并设计了启发式算法求解.基于双向随机退化Nguyen&Dupuis路网模拟了10类用户同时出行的均衡状态.分析了均衡时异质用户行程时间期望、标准差和累积到达时间感知价值等参数与用户风险态度和路网退化程度之间的关系.研究发现行程时间期望、标准差与其用户类别并无显著差异;行程时间标准差与退化路网随机性正相关,但未发现行程时间期望与路网随机性相关.用户累积到达时间感知价值随风险态度增加而递减,即保守用户感受到的累积到达时间价值更小.因此在相同的引导策略下,保守用户的广义效用函数可能变化更大,政策引导效果可能更加显著.

A Continuous Heterogeneous Bounded Rational User Division and Random User Equilibrium Model

In order to study the joint influence of the risk attitude and the road network degradation on user behavior,a user division model and a multi-class user random equilibrium model are introduced.Then a heuristic algorithm is introduced to solve the problem.Finally a numerical experiment with a bidirectional random degradation Nguyen&Dupuis network is conducted to the modle.It is showed that there is no significant difference between the expectation and the standard deviation of travel time among heterogeneous users.And there is a positive relationship between the standard deviation of travel time and the randomness of the network,It is also found that the cumulative time-of-arrival perceived value decreases monotonically with the increase of risk attitude coefficient,that is,the cumulative time-ofarrival perceived value of conservative users is smaller than others.