随机路网多时段随机期望——超额用户均衡模型

Multiple periods stochastic mean-excess user equilibrium model in stochastic network

为更准确描述随机路网环境下出行者规避风险的择路行为,对期望—超额交通分配模型进行扩展,提出一种供应与需求双重不确定条件下多时段随机期望—超额用户平衡模型。在模型中,不仅同时考虑了行程时间随机变化条件下的可靠性和不可靠性,而且还考虑了出行者对行程时间的估计误差和需求的时变性。推导了需求服从对数正态分布和路段通行能力服从贝塔分布条件下期望—超额行程时间的解析表达式。以此为基础,建立起用等价变分不等式表示的平衡模型。研究结果表明:提出的模型是可行的、有效的;随着需求波动程度以及路段通行能力退化程度的提高,高峰时段和非高峰时段的期望最小理解期望—超额行程时间都将随之增大,部分高峰时段的出行者会转移至非高峰时段。

In order to reflect travelers＇ risk aversive route choice behaviors in a more accurate manner in a stochastic road network,by means of extending the mean-excess traffic equilibrium model, this paper proposed a multiple periods stochastic mean-excess user equilibrium model with double uncertainty in supply and demand, in which both the reliability and the unreliability under travel time variability, and the effects of travelers＇ perception errors on travel time as well as the time- varying feature of traffic demand were taken into account simultaneously. Analytically expression of the mean-excess travel time when traffic demand followed log-normal distribution and link capacity followed Beta distribution was derived. Based on it, the equilibrium model was founded, in which the equilibrium conditions were formulated as an equivalent variational inequality problem. Results show that, the proposed model is valid and feasible; with the increase in demand variation level and degra- dation degree of link capacity, the mean minimum perceived mean-excess travel times in both peak period and off-peak period will become greater accordingly, as well as, part peak period travelers will shift to off-peak period.