基于状态依赖排队模型的车辆疏散问题研究

Vehicle evacuation problem based on state dependent queue model

利用状态依赖排队模型对车辆疏散过程进行仿真模拟.将车辆的选路行为划分成理性选路、随机选路和逆向行驶三类.研究了车辆在不同选路策略、不同的出口放行策略以及部分或全部交叉路口实施交通管制策略情况下对疏散时间及其分布产生的影响.研究结果表明:存在一个最优的出口车辆放行数量;理性选路的车辆越多,疏散时间越短;实施交通管制的交叉口数量越多,疏散时间越短.在中型路网中,若有部分车辆对路网不熟悉而发生了逆行现象,则车辆的平均疏散时间会显著提高,此时应该加强交通管制.研究结果可对紧急情况下,认识交通疏散现象和对交通管理方式提供建议.

State dependent queue model is used to simulate the vehicle evacuation process. Routing strategy of vehicles is classified into three categories： rational routing, random routing and reverse routing. Different routing strategies, varied exit discharge rate and traffic control on all or partial intersections have different im- pacts on clear time and its distribution. Simulation results show that： there is an optimal vehicle discharge rate at exit; the more vehicles choosing rational routing strategy, the shorter the clear time; the more intersections under traffic control, the shorter clear time. In a medium size network, the average evacuation time is increased significantly if there are some vehicles not familiar with the network. In this case, traffic control should be im- plemented. The paper tries to help in understanding more the evacuation phenomena and provides suggestions on traffic management.