多方式诱导信息对通勤者出行链的动态影响模型

为研究多方式诱导下通勤者的动态选择行为,采用效用理论和多阶段决策方法,以出行链感知效用最大化为目标,建立了通勤者单日内以家为起点和终点的出行链选择行为动态规划模型,并利用Dijkstra算法求解.将出行链分割为相互联系的单次出行,每次出行均包含出行前和出行中决策节点,在每个决策节点上实现交通信息的动态加载,并考虑通勤者对信息的信任度和学习过程,使模型可更准确地描述实际决策过程.算例结果表明:多方式诱导信息使每位通勤者的出行链实际效用平均增加0.88%;且信任度越高,通勤者获得的收益越大,越倾向于通过缩短出行链长度,选择地铁或停车换乘方式以规避交通拥堵造成的效用损失. 

Modeling the Dynamic Impacts of Multi-modal Guidance Information on Commuters' Trip Chain

To study commuters' dynamic choice behavior under the multi-modal guidance information, based on utility theory and multi-stage decision method, a dynamic model was proposed to describe the choice behavior of the daily trip chain that starts and ends at home with the objective of maximizing the perceived utility of a trip chain, and a Dijkstra algorithm was used to solve the model. In the model, the trip chain was divided into some single but sequential trips, and each trip contained pre-trip and en-route decision nodes, on which the travel information was loaded dynamically. Commuters' reliance on information and the learning process were also taken into account to describe the actual decision process more accurately. Computational results show that multimodal guidance information contributes to increase the actual utility of the trip chain by an average of 0.88% for each commuter; in addition, as the reliance increases, commuters will benefit more and tend to shorten the length of trip chain as well as to choose metro or park-and-ride, in order to avoid the utility loss from traffic congestion.