Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/21120

Title: Activity-Based Travel Demand Modeling Framework FEATHERS: Sensitivity Analysis with Decision Trees
Authors: Bao, Qiong
Kochan, Bruno
Shen, Yongjun
Bellemans, Tom
Janssens, Davy
Wets, Geert
Issue Date: 2016
Publisher: Transportation Research Board
Series/Report: Transportation Research Record: Journal of the Transportation Research Board
Series/Report no.: 2564
Abstract: The technique of decision trees is commonly applied in activity-based travel demand modeling. It owns the strength of representing the full complexity of interactions between different variables. However, this complexity on the other hand often hinders an interpretation in terms of the relative impacts of these variables on the activity travel choice. In this study, a sensitivity analysis is performed on decision trees in FEATHERS, an activity-based micro-simulation modeling framework, with the purpose of quantitatively measuring the relative impact of input variables (condition variables) involved in the given decision trees on the choice variable. Both of the local and global sensitivity analysis approaches are investigated: i) a one-at-a-time approach which predicts the choice frequency distribution by varying selected input condition variables one after another, and keeping all other variables as observed; and ii) the improved Sobol’ method which evaluates the effect of an input variable while all other variables are varied as well. By applying these two approaches to two representative decision trees concerning work related activity (i.e., commute trip) choice and transport mode choice for work-related activities in the FEATHERS framework, consistent results about the key input variables for these two decision trees are derived, and some extra insights are gained from each of these two approaches.
URI: http://hdl.handle.net/1942/21120
Link to publication: http://pubsindex.trb.org/view/2016/C/1392574
DOI: 10.3141/2564-10
ISI #: 000392257600011
ISSN: 0361-1981
Category: A1
Type: Journal Contribution
Validation: ecoom, 2018
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
Published version691.07 kBAdobe PDF
Peer-reviewed author version186.5 kBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.