考虑信息诱导的随机均衡交通流逐日演化模型

A Day-to-day Evolution Model of Stochastic Equilibrium Traffic Considering Information Guidance

在交通诱导系统向出行者发布一致性路网信息的情况下,假设出行者的路径选择行为随历史经验和交通信息的更新而不断调整,考虑出行者的路径选择行为和对诱导信息的信任决策,提出一个交通流逐日演化模型。在该模型中,根据信任决策不同,将出行者分为两类,即信任路径诱导信息的出行者和不信任路径诱导信息的出行者。其中,信任诱导信息的出行者比例由诱导信息为出行者带来的平均出行阻抗节省量决定,节省量越多,信任诱导信息的出行者比例越大,通常选择信任路径诱导信息的出行者对出行阻抗的认知更为准确。首先对演化模型进行了稳定性分析,证明了动态系统存在不动点,该不动点稳定且等价于随机用户均衡态。然后对演化模型进行数值分析,结果表明:不同的参数取值会影响演化收敛的速度和最终平衡态,说明出行者的路径选择行为和信息信任决策与交通流演化的收敛性和稳定性之间有较强的联动关系;在复杂多变的交通网络中,实时交通诱导信息可以帮助出行者降低行为惯性的负效应,减少感知偏差,为出行者提供有效的决策参考,但过高的信息信任比例会导致出行者的二次集聚,发生交通“移堵”现象;交通流随机用户均衡态下信任和不信任路径诱导信息的出行者比率通常保持在一个稳定值。

In the case that the traffic guidance system releases consistent traffic network information to travelers, assuming that the route choice behaviors of travelers are constantly adjusted with the historical experience and the updated traffic information, considering travelers’ route choice behaviors and trust decision-making on the induced information, a day-to-day evolution model of traffic is proposed. In this model, travelers are divided into 2 groups according to different trust decision-makings, i.e., travelers who trust route guidance information and travelers who don’t trust route guidance information. Among them, the proportion of travelers who trust the guidance information is determined by the average travel impedance savings that the guidance information brings to travelers. The greater the savings, the greater the proportion of travelers who trust the guidance information. Usually, travelers who choose to trust the route guidance information have more accurate perception of travel resistance. First, the stability analysis of the evolution model proves that the dynamic system has a fixed point, which is stable and equivalent to random user equilibrium state. Then, the evolution model is numerically analyzed. The result shows that(1) Different values of parameters will affect the evolution convergence speed and the final equilibrium state, indicating that there is a strong linkage among travelers’ routing behavior, information trust decision-making, convergence and stability of traffic flow evolution.(2) In the complex and changeable traffic network, real-time traffic guidance information can help travelers reduce the negative effects of behavioral inertia, reduce perception bias, and provide travelers with effective decision-making references. However, an excessively high information trust ratio will lead to the second gathering of travelers and the phenomenon of traffic jam “shifting”.(3) The proportion of travelers who trust and distrust the route guidance information under the stochastic user equilibrium state of traffic flow is usually kept at a stable value.