COVID-19 mobility decision support

A Grafana-style observability view of the community-engaged COVID-19 mobility decision-support system. Minnesota, March 2020 to June 2021.

Paper PDF Article (AGILE 2022) Project and data All publications
SafeGraph long-duration visits
weekly visit counts by dwell time, Minnesota
Mar 2020 to Jun 2021 69 weeks Measured data
Full-service, trough
-
Bars, trough
-
Limited-service, trough
-
SafeGraph sampling
~5%
294,014 devices, of MN population
Long-duration visits (over 20 min), indexed to pre-pandemic baseline = 100
Full-service (solid) Bars (dashed) Limited-service (dotted)
Long-duration visits by dwell time, per venue (the paper's Figures 8 to 10)
under 5 min 5 to 20 21 to 60 61 to 240 over 240
Full-service restaurants
Limited-service restaurants
Bars

About this dashboard

This panel reframes the community-engaged decision-support system from Sharma et al. (AGILE: GIScience Series, 2022) as an observability dashboard. The system was built with policymakers in public health, economic management, transportation, and transit to track long-duration visits to high-risk venue categories from aggregated, privacy-protected SafeGraph mobile-device data.

The charts plot the project's measured weekly visit counts to bars, full-service restaurants, and limited-service restaurants in Minnesota, broken out by dwell-time bucket (under 5, 5 to 20, 21 to 60, 61 to 240, and over 240 minutes), from March 2020 to June 2021. Markers show the major Minnesota policy interventions (the March 27, 2020 stay-at-home order, the June 2020 reopening, the November 2020 dining shutdown, and the January 2021 reopening). The per-venue panels mirror Figures 8 to 10 of the paper; the overview indexes the long-duration visits (over 20 minutes) of each venue to its own pre-pandemic baseline so the three are comparable on one axis.

Data: the project's weekly SafeGraph long-duration-visit reports, 07/16/2021 release (69 weeks, 03/02/2020 to 06/28/2021). Source spreadsheets: full-service, limited-service, bars. SafeGraph covers about 5 percent of the Minnesota population (294,014 devices); see the paper for the sampling-bias analysis. Citation: Arun Sharma, Majid Farhadloo, Yan Li, Jayant Gupta, Aditya Kulkarni, Shashi Shekhar, "Understanding COVID-19 Effects on Mobility: A Community-Engaged Approach," AGILE: GIScience Series 3:14 (2022), CC-BY 4.0.