ROUTE CHOICE BEHAVIOUR ON CONGESTED MOTORWAY NETWORKS: A STATED PREFERENCE EXPERIMENT

The objective of this study is to analyze motorists' route choice in congested motorway networks. This analysis should result in a route choice model that can be incorporated in a simulation model. This simulation model will be used to determine the effectiveness of alternative designs of road-side route guidance systems with variable message signs. Control strategies have also to be evaluated. Using a stated preference experiment among about 200 employees of large organizations located in a congested corridor a number of route choice models of the logit type have been established. Route choice factors included are route length in kilometers and delay due to congestion. At the aggregate level it is shown that the simple model that is linear in the parameters and the attributes has the best performance. There is no need to use more complex models; inclusion of non- linearities and interaction effects did not improve the statistical performance. The model shows that route choice behaviour in the context investigated is different from what might have been expected using a simple time- minimizing rule based on free-flow speeds. Segmentation of the population based on socio-economic characteristics of the respondents did not improve the performance of the model. Almost every class of respondents had almost an identical aggregate route choice behaviour. In addition, a route choice model has been estimated for every individual respondent. These individual models had an identical structure to those at the aggregate level. The main finding was that individual travellers exhibit a rather large variation in the model parameters; this variation which could not be related to exogenous factors in this study, should be taken into account explicitly when modelling the effects of route guidance systems. (A)