Abstract:

Travellers’ route choice behaviour often has the characteristics of bounded rationality such as regret avoidance under the situation that travel time is uncertain. The intensity of regret caused by the result of choice is closely related to travel habits. In this paper, travellers’ regret avoidance psychology and travel habits are considered simultaneously in the traffic assignment problem. Travellers are classified according to the familiarities of routes between origin-destination pairs. Then, a multi-class stochastic user equilibrium model is established based on the regret theory considering route familiarity. The equivalent variational inequality (VI) is presented, and the solution existence is shown. The result of numerical examples shows that the proportion of users with familiar routes and the difference of their familiarities to each route will significantly affect the equilibrium traffic flow. The new model proposed in this paper can provide theoretical support for travel behaviour analysis, network traffic flow analysis, and traffic control strategy formulation from the perspective of behavioural psychology.

Key words:

traffic assignment, regret theory, decision inertia, route choice, multi-class users

摘要：

出行时间不确定情境下，人们的路径选择行为往往表现出后悔规避等有限理性特征。由选择结果引起的后悔情绪，其强度与人们的出行习惯关系密切。在交通流分配问题中同时考虑出行者的后悔规避心理与出行行为习惯，将出行者按照起讫点间不同路径的熟悉程度进行分类，构建了后悔视角下考虑出行者路径熟悉度的多用户随机用户均衡模型。给出了与模型等价的变分不等式，并证明了解的存在性。算例结果表明：具有熟悉型路径的用户比例以及用户对路径的熟悉度差异显著影响网络均衡流量分布。提出的新模型能够从行为心理学角度为出行行为分析、网络流量构成解析、交通管控策略制定等方面提供理论支持。

关键词:

交通流分配, 后悔理论, 决策惯性, 路径选择, 多用户