COVID-19 Simulator: An interactive online tool for management of COVID-19 pandemic

Audience Level: 
All
Institutional Level: 
Higher Ed
Streamed: 
Streamed
Special Session: 
MERLOT Award Winner
Abstract: 

The COVID-19 Simulator is an interactive publicly available tool, which is calibrated to reproduce historical trends in daily reported cases and deaths and is updated weekly as new data and evidence arise. The Policy Simulator influenced the timing of the lifting of restrictions in multiple states after the lockdown in March-April 2020, and its findings are regularly cited by media outlets, including NPR and Newsweek. Since May 2020, the CDC has incorporated our model outputs in its weekly COVID-19 forecasts. The tool also predicts the doubling time of COVID-19 cases in each county, which tells how how fast COVID-19 is spreading in a county.

Extended Abstract: 

The COVID-19 Simulator is an interactive publicly available tool, which is calibrated to reproduce historical trends in daily reported cases and deaths and is updated weekly as new data and evidence arise. The Policy Simulator influenced the timing of the lifting of restrictions in multiple states after the lockdown in March-April 2020, and its findings are regularly cited by media outlets, including NPR and Newsweek. Since May 2020, the CDC has incorporated our model outputs in its weekly COVID-19 forecasts.

The COVID-19 tool was expanded to include COVID-19 Outbreak Tool, which predicts the doubling time of COVID-19 cases in each county, which tells how how fast COVID-19 is spreading in a county. Using a machine learning approach, the tool estimates the doubling time of COVID-19 cases in each county by accounting for reported COVID-19 cases, COVID-19 deaths, face mask mandate, social distancing policies, social vulnerability index of each county (e.g. income level, employment rate), as well as the daily COVID-19 test and vaccination ratios of each county. The tool is updated at least once a week.

Conference Track: 
Instructional Technologies and Tools
Session Type: 
Discovery Session Asynchronous
Intended Audience: 
All Attendees