A hybrid approach to combine fuzziness and randomness in travel choice prediction

Abstract

Several approaches have been developed in order to deal with uncertainty, in the prediction of travellers’ choices. Uncertainty almost always affects travel alternatives in several different choice contexts. However, the way in which this uncertainty affects choice options may consistently vary. Two main types of uncertainty can be identified: randomness and fuzziness. This paper discusses the use of a hybrid approach for choice modelling of travellers’ behaviour in choice contexts that involve conditions of high variability and uncertainty. The aim of the work is to implement a hybrid methodology, based on the use of a two-term hybrid utility, which allows taking into account, at the same time, the effects of both random and fuzzy uncertainty on travellers’ choice behaviour. An application of the proposed methodology to a transit choice context is then presented.