DEALING WITH CAPACITY CONSTRAINTS WITHIN STOCHASTIC USER EQUILIBRIUM ASSIGNMENT MODELS

One of the challenges of traffic assignment models is to properly take into account capacity constraint problems in order to avoid traffic propagation through bottlenecks. This allows better evaluations of travel times between the origin and destination (O/D) couples of the study area, adequate computations of the performances of the traffic network and design and/or simulation of traffic schemes oriented to 'protect' city centers from excessive traffic flows. The analysts have made most of the modeling efforts related to capacity constrained assignment on dynamic network loading procedures or on static, but deterministic, assignment models. In this paper a static, stochastic and behaviorally consistent assignment model is presented, as well as its application to a real-size network.