In the past, route choice analyses for motorised modes have been carried out along two main approaches. In one approach travellers are askes for their reasons or motivations in selectiong routes in a network, resulting in an ordered list of choice factors. Travel time usually arises as the most important of these factors. The revealed preference appraoch, on the other hand, allows estimation of the relative importance of the choice factors by comparing actual chosen and non-chosen alternatives. Both approaches have drawbacks in the context of route choice. This was the impetus for exploring the potentials of stated preference methods. This report describes an investigation into the usefulness of stated preference survey techniques in modelling route choice behavior, in particular that of bicyclists. Special attention has been devoted to the development of the survey instrument. This refers to two aspects: first, the way how various elements of bicycle route choice such as relevant choice factors, route alternatives, and choice situations are described and presented to subjects in the contect of hypothetical choices. Secondly, it refers to the way how preferences of subjects with respect to alternative routes are measured and analysed in such a context. This study was carried out with over one hundred bicycle commuters in the City of Delft. The relative importance of travel time, surface quality, traffic level, and cycling facilities was estimated for the daily home-to-work, using a block design of twenty-seven hypothetical routes. The methods of factor descriptor, grouping of alternatives, measuring preference, and estimation were varied systematically to test their influence on the results.
Abstract