A Stochastic User Equilibrium Assignment Model under Stochastic Demand and Supply Following Lognormal Distributions

抄録

To construct a traffic network that is resilient to natural and man-made disasters, it is necessary to assess the reliability of the network. This article proposes a stochastic user equilibrium assignment model under stochastic origin-destination (OD) demand and link capacity following lognormal distributions. The nonnegative property of the lognormal distribution makes it possible to assume any degraded state of links, including disruption or reduction of lane(s), by varying the stochastic link capacities. This model assigns the OD demand to a road network based on the stochastic user equilibrium principle. Thus, the model can compare the link (or path) flows and link (or path) travel costs in the normal state with those in the degraded state, and is formulated as a logit-based fixed-point problem. We conducted numerical experiments to compare the assigned flow and travel cost in the normal state with those in the degraded state.