Traditional multimodal travel choice modeling approaches, applicable to single-trip data, distinguish access and egress but make no difference between travelers’ home and activity addresses. It is hypothesized that the differences between home end and activity end with respect to a number of travel-related factors are relevant in the travel choice process and therefore need to be accounted for correctly in multimodal travel choice models. It is shown empirically that differences indeed exist in availability, knowledge, and use of multimodal trip alternatives. In the analysis, revealed-preference data for interurban multimodal train trips in the Netherlands are used. A new home–activity modeling approach is presented: the home end and the activity end are explicitly taken into account in the utility specification, and all attributes are direction free. To illustrate the approach, a generalized nested logit model, suited to capture correlations caused by home-end and activity-end feeder mode types, is applied to the multimodal train trip data. The existence of differences in travelers’ valuation of home-end and activity-end characteristics was found. More specifically, home-end characteristics play a more important role in planning multimodal train trips than activity-end characteristics.
Home–Activity Approach to Multimodal Travel Choice Modeling
Transportation Research Record: Journal of the Transportation Research Board
Transportation Research Record: Journal of the Transportation Research Board ; 1985 , 1 ; 180-187
2006-01-01
Article (Journal)
Electronic Resource
English
Home-Activity Approach to Multimodal Travel Choice Modeling
Transportation Research Record | 2006
|Home-Activity Approach to Multimodal Travel Choice Modeling
Online Contents | 2006
|Activity choice analysis time allocation and disaggregate travel demand modeling.
DSpace@MIT | 1976
|Destination Choice in Daily Activity Travel
Transportation Research Record | 2008
|Travel choice behavior modeling : special issue
TIBKAT | 2008
|