Equilibrium Assignment Model with Uncertainties in Traffic Demands

In this paper, we present an equilibrium traffic assignment model considering uncertainties in traffic demands. The link and route travel time distributions are derived based on the assumption that OD traffic demand follows a log-normal distribution. We postulate that travelers can acquire the variability of route travel times from past experiences and factor such variability into their route choice considerations in the form of mean route travel time. Furthermore, all travelers want to minimize their mean route travel times. We formulate the assignment problem as a variational inequality, which can be solved by a route-based heuristic solution algorithm. Some numerical studies on a small test road network are carried out to validate the proposed model and algorithm, at the same time, some reasonable results are obtained.