A Modification of Multiple Discrete-Continuous (MDC) Choice Model to Consider Nonmonotonic Preference in Episode-Level Time-Use Behaviors

The multiple discrete-continuous extreme value model with ordered preferences (MDCEV-OP) has broad prospects in activity-based modeling (ABM) since it can model episode-level time-use decisions and ensure a logical prediction across different episodes of an activity. However, the current MDCEV-OP fr...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengyi Wang, Xin Ye, Ke Wang
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:Journal of Advanced Transportation
Online Access:http://dx.doi.org/10.1155/atr/7114605
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The multiple discrete-continuous extreme value model with ordered preferences (MDCEV-OP) has broad prospects in activity-based modeling (ABM) since it can model episode-level time-use decisions and ensure a logical prediction across different episodes of an activity. However, the current MDCEV-OP framework assumes a monotonically increasing utility function for each episode alternative, which fails to accommodate potential nonmonotonic preference in episode-level time consumption. In this paper, we modify the traditional MDCEV-OP model by adding a baseline marginal utility parameter, making the model more flexible to reflect the potential nonmonotonic preference in episode-level time-use behaviors, as well as ensuring the logically consistent prediction as in the traditional model. To our knowledge, it is the first time to develop an episode-level MDCEV model that considers nonmonotonic preference. The new MDCEV-OP model was applied to analyze the episode-level time-use pattern of noncommuters in Shanghai, China. The empirical results show that the new model provides plausible explanations for nonmonotonic preference in episode-level time-use behaviors and outperforms the traditional model both in data fitting and forecasting performance.
ISSN:2042-3195