経路選択行動のday-to-dayダイナミクスと交通ネットワーク均衡の形成プロセス

概要

論文の詳細を見る
In this study, we assume that each driver under day-to-day dynamic transportation circumstances chooses a route based on Bayesian learning, and develop a day-to-day dynamical model of network flow. It is found in this model that the driver with Bayesian learning chooses the route which has the minimum travel time the most frequently. Furthermore, we find that an equilibrium point of the day-to-day dynamical model is identical to the Wardrop's equilibrium, and the Wardrop's equilibrium is globally asymptotically stable if initial recognition among drivers is dispersed widely, and the day-to-day dynamical system always converges to the Wordrop's equilibrium.