Transportation demand models and traffic simulations, particularly in the realm of innovative activity-based models, necessitate comprehensive demographic-activity-travel data. Although cell phone data holds the potential to provide detailed insights into activity-travel patterns, transportation planners often resort to aggregate data, constrained by limited socio-demographic attributes because of stringent privacy protection policies. This creates a significant challenge: how to reconstruct detailed traveler profiles from such aggregate data. In this paper, we propose a novel data-driven integration framework, DATG, based on improved Markov and XGBoost classification models, with which we are able to reconstruct travelers with diverse socio-demographic attributes as well as their one-day activity-travel information. We test our model on a real cell phone data set and validate the proposed model by relying on travel survey data. Our framework is able to provide fundamental data support for agent-based models, activity-based models, and transportation planning.
Data-Driven Framework for Generating Travelers with Demographic-Activity-Travel Information
Transportation Research Record: Journal of the Transportation Research Board
Transportation Research Record: Journal of the Transportation Research Board ; 2678 , 10 ; 1-13
2024-02-02
Article (Journal)
Electronic Resource
English
Analyzing Travelers’ Intention to Accept Travel Information
Transportation Research Record | 2010
|Travel Time Reliability with Risk-Sensitive Travelers
Transportation Research Record | 2002
|Travel Time Reliability with Risk-Sensitive Travelers
Online Contents | 2002
|Travel time reliability with boundedly rational travelers
Taylor & Francis Verlag | 2018
|Travel Time Reliability with Risk-Sensitive Travelers
British Library Conference Proceedings | 2002
|