This paper will mainly deal with binary choice and ANN models of route choice and driver compliance with traffic information. This work is part of a study aimed at incorporating driver compliance models within traffic simulation tools to improve their accuracy in evaluating Intelligent Transport Systems (ITS) applications. It is worthwhile to note that a comparative evaluation of the modelling approaches of driver route choice and compliance with traffic information, based on the same data set, has not been reported in the literature. For this study, neural techniques were developed to predict drivers' compliance with travel information and compared against the traditional utility models using the same data set of commuter behaviour. (a) For the covering entry of this conference, please see ITRD abstract no. E214666.
Samenvatting