2025 年 16 巻 論文ID: PP4156
In this paper, a method for modeling user behavior in a tolled road network is proposed. The effective travel time is defined as the mean travel time plus the travel time variance multiplied by a risk aversion parameter. The sources of the stochasticity of travel time are O-D demand and link capacity which follow a lognormal distribution. Each user has a maximum effective travel time they can spend traveling between origin and destination nodes, which is determined by an indifference function between toll and the effective travel time. The effective travel time surplus is defined as the maximum effective travel time minus the perceived effective travel time, i.e., each user has a perception error. Each user chooses a path so that the surplus is maximized. This study proposes an effective travel time surplus maximization problem, which is formulated as a stochastic user equilibrium model under stochastic demand and link capacity.