多用户类多准则交通分配的势博弈与拥挤定价

POTENTIAL GAME OF MULTI-CLASS,MULTI-CRITERIA TRAFFIC ASSIGNMENT AND CONGESTION PRICING

交通管理者在解决路网拥挤问题时,并不知道出行者的出行效用,同时管理者难以对出行者的路径选择行为做出准确的观测.运用势博弈理论分析多用户类多准则交通行为的演化过程,得到了固定需求和弹性需求情形下的可容许动态(一种刻画出行者通过转换路径增加当前效用的近似调整行为的演化动态),证明当路段时间函数和逆需求函数为严格单调、连续、可微时,所对应的交通分配是势博弈问题的惟一Nash均衡点.进一步研究了固定需求下的可变拥挤道路收费问题,得到了在当前系统状态下实现系统最优交通分配的拥挤收费水平.

Managers of a transportation network don＇t know the utilities of road users when resolving the urban traffic and they have a limited ability to observe road users＇ path choice behavior. In this paper, the potential game theory is used to study the evolution of multicriteria traffic assignment with multiple user classes. An admissible dynamic, i.e., a specific evolutionary dynamic which models the behavior as a myopic adjustment process of switching paths for current utility improvement, in the case of either fixed demand or elastic demand, is derived. It is shown that the corresponding traffic assignment is the Nash equilibrium of a potential game if both link travel time function and inverse demand function are strictly monotonic, continuous and differentiable. Furthermore, a variable congestion pricing is given, which can drive the multi-class multi-criteria traffic assignment at current system status to be optimal.