Mobile Device Data Analytics for Next-Generation Traffic Management

Research Team: Jane Macfarlane (lead), Anthony D Patire, Kanaad Deodhar, and Colin Laurence

UC Campus(es): UC Berkeley

Problem Statement: The availability of detailed location data from personal mobile devices and fleet transponders has increased significantly in recent years. Carrying rich, detailed information on travel patterns, this data has already seen broad adoption in the transportation planning industry, for example,informing corridor studies and travel demand models. This report outlines an architecture and computational framework for the ingestion, processing, and analysis of raw location data, that can be used to transform GPS points from mobile devices into actionable insights about the transportation network.This method was used to generate traffic data for the San Francisco Bay Area and applied to a recent bridge closure to examine changes in local traffic patterns.

Project Description: The core focus of this work is to provide tools for research in transportation planning by enabling organizations and researchers across California to effectively take advantage of location data. For example, researchers will be able to better understand regional travel flows and understand the congestion impacts of emergencies and disasters, two applications which are explored in this report. Though this framework has been developed with proprietary location data and roadway network information provided by HERE Technologies, it can be generalized to operate on any type of geospatially referenced data and open-source maps, e.g., OpenStreetMap.

Status: Completed

Budget: $76,083

Project Partner(s): Caltrans

Report(s):
Download Report