Demand-driven traffic assignment problem based on travel time reliability

A novel user equilibrium traffic assignment model based on travel time reliability is presented in view of day-to-day demand fluctuation. Because of daily demand variations, path travel times are not constants, and so they can be viewed as random variables. Assuming that travelers are able to learn the variation of path travel time from experience, a demand-driven user equilibrium principle based on travel time reliability is proposed to characterize travelers' path choice behavior under uncertainty in travel times caused by demand variation. This principle can be formulated as a variational inequality (VI) problem in terms of path flows. The existence of at least one solution for the VI problem, for which a heuristic solution algorithm is adopted, is rigorously proved. Numerical examples are used to illustrate the applications of the proposed model and the solution algorithm.