基于期望-超额出行时间的道路系统最优均衡模型

为研究随机事件扰动下出行者的择路行为对交通分配的影响，同时考虑供需条件的随机变化，以期望-超额出行时间为出行者择路依据，利用边际成本收费原理，推导了边际成本收费值计算公式，建立用等价变分不等式表示的系统最优交通分配模型，并利用自适应投影收缩算法进行求解.算例表明:当OD需求系数为1.0、路段能力退化系数为0.5时，路径1边际成本收费值分别比使用期望出行时间和出行时间预算为择路依据时增加了11.27%和3.58%；当出行时间可靠度为0.9时，路径1边际成本收费值分别比使用期望出行时间和出行时间预算作为择路依据时增加了20.22%和4.30%. 

Road System Optimum Equilibrium Model Based on Mean-Excess Travel Time

In order to study the effects on traffic assignment of travelers' route choice behaviors under random perturbations, the mean-excess travel time (METT) was assumed to describe travelers' route choice criteria. According to the theory of marginal cost pricing, the calculation formula for marginal cost pricing was derived under stochastic supply and demand. Then, a traffic assignment system equilibrium model was built and formulated as an equivalent variational inequality,and a self-adaptive projection and contraction method was used to solve the model. The results of numerical experiments show that when the ratio of origin-destination (OD) demand coefficient was 1.0 and the ratio of degradation of link capacity was 0.5, the marginal cost toll of route 1, compared with those by use of mean travel time and travel time budget to choose paths, was increased by 11.27% and 3.58%, respectively. When the reliability of travel time was 0.9, the marginal cost toll of route 1 were found to be 20.22% and 4.30% higher than those using the mean travel time and travel time budget as a basis for route choice, respectively.