Traffic Information and Learning in Day-to-Day Route Choice

Abstract

The aim of the research is to gain insight into the impact of traffic information on day-to-day route choice. The effects of both en-route information and ex-post information are studied. Special attention is given to three other important aspects in day-to-day route choice: learning from experience and information, habits and the effect of uncertainty / reliability. A large experiment was set up and a discrete choice model for panel data was estimated on the resulting data. Conclusions: • The more elaborate traffic information that was provided, the higher the travel time savings that were realized. • Travellers chose less reliable routes more often when they were provided with traffic information. • Whereas travellers who did not receive en-route traffic information based their expected travel time mainly on the most recent experience, travellers with en-route information used more previous experiences. • The role of past choices (used to model implicit learning and habit formation) compared to the role of the expected travel time (used to model explicit learning) was found to be relatively large in the route choice process.