A hybrid model for en-route driver behavior under real-time information

Abstract

The driver en-route route choice problem under real-time information is characterized by subjectively interpreted and/or linguistically expressed data on nonquantitative variables, as well as limited data on some quantitative variables. This paper proposes a hybrid fuzzy logit model to enable the treatment of nonquantitative and quantitative data simultaneously in a single framework to address the problem. Synergistically, its computational efficiency enables real-time deployment in a traffic control framework. The study highlights the appropriateness of using fuzzy variables to address recently identified en-route driver behavioral characteristics under information provision.