A revealed-preference study of behavioural impacts of real-time traffic information

In the present study, we investigate the impact of real-time traffic information on traveller behaviour by using repeated day-to-day revealed-preference (RP) observations from a reward experiment. We estimate a trip scheduling model of morning peak behaviour that allows us to determine the impact of traffic information on travel behaviour. Specifically, we distinguish between the marginal impact of expected travel times versus that of deviations from this expectation upon user behaviour. We find that participants to the experiment who were given access to a smart-phone displaying real-time traffic information react to the daily variations in travel times stronger than they had done in the pre-trial period. This we interpret as evidence that provision of traffic information indeed affects behaviour. However, we also find that participants who were not given the smart-phone also respond to daily variations in travel times – suggesting that these drivers use other sources of information to help their trip planning, or were better capable of processing information available into travel time predictions