Modeling discrete choice behavior based on explicit information integration and its application to the route choice problem

Two-stage framework for modeling discrete choice behavior in the presence of new information is presented. It is based on explicit information integration at the first stage, and a decision process based on the integrated information at the second stage. The first stage is conducted using an integration scheme which is source- and content-dependent, and the second stage is implemented using an approximate-reasoning-based choice model. The existence of an explicit information integration stage provides the advantage of improving the calibration procedure of the choice model by facilitating a two-phase calibration process in which both rules and membership functions are calibrated. A simple case study demonstrates the applicability of the approach for modeling human route choice decisions in transportation networks when updated online traffic information is available to drivers. The case study also provides interesting insights on the trade-offs between the number of rules and the importance of generating good membership functions. The suggested decision model is roughly compared to traditional random utility model, and provides favourable results