Identifying High-Risk Workspaces during COVID-19 using Machine Learning

Authors

  • Lex Drennan Steelcase
  • Matthew Chesser University of Michigan
  • Jorge Lozano Steelcase
  • Erin Carrier Grand Valley State University

DOI:

https://doi.org/10.32473/flairs.v34i1.128484

Keywords:

machine learning, COVID-19, risk detection

Abstract

The COVID-19 pandemic has wreaked havoc worldwide, on both public health and the worldwide economy. While necessary, quarantine and social distancing requirements have left many companies unable to reopen their offices in a safe manner. We present a model capable of identifying workspaces at high risk for COVID-19 disease transmission and illustrate how existing techniques for quantifying uncertainty in machine learning can be applied to assess the reliability of these predictions. This model is developed using a dataset created by leveraging historical sales data and detailed product information, and it is in the process of being utilized to identify customers to whom to reach out to facilitate the retrofitting of workspaces to support a safe return to the office.

Downloads

Published

2021-04-18

How to Cite

Drennan, L., Chesser, M., Lozano, J., & Carrier, E. (2021). Identifying High-Risk Workspaces during COVID-19 using Machine Learning. The International FLAIRS Conference Proceedings, 34. https://doi.org/10.32473/flairs.v34i1.128484

Issue

Section

Main Track Proceedings