From Macro to Microscopic Trip Generation: Representing Heterogeneous Travel Behavior
Rolf Moeckel*, 1, Leta Huntsinger2, Rick Donnelly3
Identifiers and Pagination:Year: 2017
First Page: 31
Last Page: 43
Publisher Id: TOTJ-11-31
Article History:Received Date: 19/09/2016
Revision Received Date: 18/12/2016
Acceptance Date: 06/01/2017
Electronic publication date: 23/03/2017
Collection year: 2017
open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
In four-step travel demand models, average trip generation rates are traditionally applied to static household type definitions. In reality, however, trip generation is more heterogeneous with some households making no trips and other households making more than a dozen trips, even if they are of the same household type.
This paper aims at improving trip-generation methods without jumping all the way to an activity-based model, which is a very costly form of modeling travel demand both in terms of development and computer processing time.
Two fundamental improvements in trip generation are presented in this paper. First, the definition of household types, which traditionally is based on professional judgment rather than science, is revised to optimally reflect trip generation differences between the household types. For this purpose, over 67 million definitions of household types were analyzed econometrically in a Big-Data exercise. Secondly, a microscopic trip generation module was developed that specifies trip generation individually for every household.
This new module allows representing the heterogeneity in trip generation found in reality, with the ability to maintain all household attributes for subsequent models. Even though the following steps in a trip-based model used in this research remained unchanged, the model was improved by using microscopic trip generation. Mode-specific constants were reduced by 9%, and the Root Mean Square Error of the assignment validation improved by 7%.