To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2005.02.026 Byline: Vincent Henn (a), Michele Ottomanelli (b) Keywords: Traffic assignment; Route choice; Uncertainty; Possibility theory; Expected utility theory; Fuzzy logic Abstract: In order to design effective advanced traffic information systems (ATIS) suitable mathematical models have to be defined to simulate the effects of information on users route choice behaviour and then to incorporate it into traffic assignment models to estimate how traffic demand loads the roads network. To face this problem it is necessary to deal with uncertainty that plays a crucial role in the users decision-making processes. To this purpose this paper first analyses how uncertainty affects users' route choice process and how traffic assignment models may take it into account. In literature route choice behaviour modelling is widely solved within the random utility theory framework but, we show in this paper that such an approach only considers one type of uncertainty. More precisely, the consideration of randomness of traffic by drivers is, for example, hardly ever represented in classical models in spite of its importance in the management of information by drivers. Starting from the presented analysis a new route choice model is also proposed to represent explicitly the uncertainty lying in users' route choice behaviour. It is based on recent developments in possibility theory which is an alternate way to probability theory in order to represent or measure uncertainty. Author Affiliation: (a) LICIT -- Ecole Nationale des Travaux Publics de l'Etat, Rue Maurice Audin, F-69518 Vaulx-en-Velin Cedex, France (b) Polytechnic University of Bari, Department of Environmental Engineering and Sustainable Development, Viale del Turismo, 8-74100 Taranto, Italy