ATIS环境下的网络交通流逐日动态演化研究

Day-to-day dynamic evolution of network traffic flow under ATIS

基于出行者逐日路径选择过程,建立出行者路径选择模型和逐日路径选择学习模型.在2条平行路径的简单路网中进行多次数值实验,结果表明:网络交通流演化是否稳定,以及能否达到随机用户均衡态,与出行者对昨日感知行程时间的依赖程度和先进的出行者信息系统市场占有率两者都有关系.当出行者对昨日实际行程时间完全依赖时,路网系统始终偏离随机用户均衡态;当出行者对昨日感知行程时间完全依赖时,出行者集中选择一条路径,并且网络交通流演化稳定.

Based on the route choice process,the route choice model is presented,and the day-to-day route choice learning models are also presented. This paper makes a lot of numerical experiments in a simple network with two parallel routes. The results show that whether the network traffic flow evolution is stable and can reach stochastic user equilibrium depends on the travelers’ reliance degree to yesterday’s perceptive travel time and ATIS market penetration. When the travelers rely on yesterday’s actual travel time completely,the system of network has always deviated from stochastic user equilibrium. When the travelers rely on yesterday’s perceptive travel time completely,travelers choose the same route collectively and network traffic flow evolution is stable.