A disaggregate logit model of intercity common carrier passenger modal choice

Abstract

The use of the logit functional form is common in traditional, aggregate intercity passenger demand models. The use of disaggregate logit models (that is, models calibrated on data pertaining to individual travellers' behaviour and derivable from explicit assumptions concerning travellers' decision-making processes) in the intercity context is surprisingly rare. The major obstacle to such usage would appear to be a general lack of the disaggregate data required to calibrate these models. The major rationale for their use, on the other hand, is the potential that they possess for achieving a more “behavioural” representation of travellers' decision-making and actions.The paper presents a primarily empirical study of the use of disaggregate choice models to represent one component of intercity travel demand, choice of mode. Using 1969 Canadian Transport Commission (CTC) survey data for the Windsor–Quebec City corridor, a disaggregate logit model of common carrier modal choice is presented and discussed. Despite its age and the lack of the auto mode, the CTC data set was used in this study because it appears to be the best disaggregate data set currently available in Canada. The paper also discusses several key issues involved in the development of this model and in its use in intercity passenger demand forecasting, including model data requirements and comparison of the model's predictive performance with traditional aggregate models. Key words: intercity passenger travel, mode choice, disaggregate logit model.