Skip to main content
Open Access Publications from the University of California

Demand Forecasting and Activity-based Mobility Modeling from Cell Phone Data


This project develops machine learning algorithms and methods for processing of cell phone location logs to generate travel behavior data. The project initially focuses on bias correction and activity inference for generating activity-based travel demand models. Inferred activity chains are used to calibrate an agent-based traffic micro-simulation for the SF Bay Area, and validated on loop detector counts.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View