Modeling drivers uncertainty in traffic assignment models: a possibility theory based approach

One of the most important issue in transportation system analysis deals with modeling of the effects of advanced traffic information systems (ATIS) on users´ route choice behaviour and how incorporate such effects in traffic assignment models. To face this problem it is necessary to deal with uncertainty that is the core of the users´ decision making processes. To this purpose we first analyse how uncertainty affects users´ route choice process and how traffic assignment models may take it into account by giving an overview of the main theoretical approaches, such as random utility theory based models. Starting from the analysis carried out, a new route choice model is 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 represent or measure uncertainty