Formulation and Solution Algorithm for Fuzzy Dynamic Traffic Assignment Model

An issue that is always important in the development of traffic assignment models is how travelers' perceptions of travel time should be modeled. Because travelers rarely have perfect knowledge of the road network or of the travel conditions, they choose routes on the basis of their perceived travel times. Traditionally, travelers' perceived travel times are treated as random variables, leading to the stochastic traffic assignment problem. However, uncertain factors are also observed in the subjective recognition of travel times by travelers, and these can be illustrated as fuzzy variables. Therefore, a fuzzy dynamic traffic assignment model that takes into account the imprecision and the uncertainties in the route choice process is proposed. By modeling the expressions of perceived travel times as fuzzy variables, this model makes possible the description of a traveler's process of choosing a route that is more accurate and realistic than those from its deterministic or stochastic counter parts. The fuzzy perceived link travel time and fuzzy perceived path travel time are defined, and a fuzzy shortest path algorithm is used to find the group of fuzzy shortest paths and to assign traffic to each of them by using the so-called C-logit method. The results of the proposed model are also compared with those from the stochastic dynamic traffic assignment model, and it is demonstrated that the impact of advanced traveler information systems on the traveler's route choice process can be readily incorporated into the proposed model.