Extracting Dynamic Urban Mobility Patterns during COVID-19 from Bluetooth Tracking Data

Authors: Jilin Hu*, , Mogahid Adam Hussein, Graduate Student, Yihong Yuan, Associate Professor, Khan Mortuza Bin Asad, Graduate Student
Topics: Urban Geography, Transportation Geography
Keywords: Bluetooth data; Human mobility; Big data; Urban Dynamics; COVID-19
Session Type: Virtual Paper
Day: 4/11/2021
Start / End Time: 8:00 AM / 9:15 AM
Room: Virtual 9
Presentation File: No File Uploaded


Bluetooth tracking data provide researchers an opportunity to understand the behavioral patterns of individual movement at a low-cost compared to traditional travel diaries and questionnaires. Bluetooth data can effectively capture intra-urban mobility patterns across street networks because of their high precision and sampling frequency. In this study, we extract hourly time series from the Bluetooth travel sensors dataset provided by the City of Austin to explore the changes in each sensor location's dynamic mobility patterns before and during COVID-19. A dynamic Time Warping (DTW) algorithm is applied to measure the similarity between these time series. The methodology of this research is three folded. First, we identified which areas in Austin, TX experienced the least and most similar patterns in human mobility between April 2019 and April 2020. Second, we extracted urban functional regions with the least and most similar mobility patterns before and during COVID-19. Third, we correlated the extracted urban functional regions with land use/land cover and points of interest (POI) data. This research will aid city officials in understanding how urban dynamics changed due to COVID-19. The research framework can be applied to other cities where similar datasets that capture intra-urban human mobility are available.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login