MODELLING INTER URBAN ROUTE CHOICE BEHAVIOUR

Abstract

This paper develops and tests innovative models of driver's route choice behavior. The approach taken in this paper is to define choice sets of paths. The idea is to transform a large number of physical routes into a smaller number of routes each representing a specific label. A labelled path is defined as the optimal physical path with respect to some criterion function. The criteria that might be relevant to route choice include travel time, distance, scenery, congestion, signposting, etc. The label functions are estimated by maximizing the proportion of observed routes included in the sets of labelled paths. This paper describes the analysis of a set of labels and their estimated functions. This modelling approach assumes a two stage decision process. In the first stage, a choice set of alternative routes between origin and destination are generated; these are the labelled paths. In the second stage the driver is selecting one of these routes on the basis of route attributes and personal and trip characteristics. A random utility theory is employed to derive the choice model for the second stage of this decision process. The general model has a nested-logit form that represents alternative assumptions about the unobserved route choice utilities; some unobserved attributes may be link-related and others label-related. The estimation results demonstrate the important effects of factors other than time and distance on driver's route choice behavior. (Author/TRRL)