Pandemic Lock-down, Isolation, and Exit Policies Based on Machine Learning Predictions

Production and Operations Management, Special Issue on Managing Pandemics: A POM Perspective

40 Pages Posted: 12 May 2020 Last revised: 4 Mar 2022

See all articles by Theodoros Evgeniou

Theodoros Evgeniou

INSEAD

Mathilde Fekom

University of Paris-Saclay

Anton Ovchinnikov

Smith School of Business - Queen's University; INSEAD - Decision Sciences

Raphael Porcher

Université Côte d'Azur - INSERM; Universite Paris Descartes - Center for Clinical Epidemiology

Camille Pouchol

MAP5, FP2M, CNRS FR 2036, Université de Paris

Nicolas Vayatis

ENS Paris-Saclay

Date Written: January 15, 2022

Abstract

In early May 2020, following social distancing measures due to COVID-19, governments consider relaxing lock-downs. We study how that might be done by utilising clinical risk predictions. We extended a standard susceptible-exposed-infected-removed (SEIR) model to account for personalised predictions of severity, defined by the risk of an individual needing intensive care if infected, and simulated differential isolation policies using COVID-19 data and estimates in France as of early May 2020.

Simulations indicated that, assuming everything else the same, an exit policy considering clinical risk predictions starting on May 11, the date chosen by the French government, could enable to immediately relax restrictions for an extra 10% (6 700 000 people) or more of the lowest-risk population, and consequently relax the restrictions on the remaining population months faster – while abiding to the current ICU capacity. Similar exit policies without risk predictions would exceed the ICU capacity by a multiple.

Sensitivity analyses showed that when the assumed percentage of severe patients among the population decreased, or the prediction model discrimination improved, or the ICU capacity increased, policies based on risk models had a greater impact on the results of epidemic simulations. At the same time, differential isolation policies require the higher risk individuals to comply with recommended restrictions. In general, our simulations demonstrated that risk prediction models can inform new personalised isolation and exit policies, which may lead to both safer and faster outcomes than what can be achieved without such models.

Note: Funding: No funding was used for this research.

Conflict of Interest: No competing interests.

Keywords: COVID-19, Epidemiology, Pandemics, Outbreaks, Epidemic Models, Machine Learning, Data Science, Prediction Models

JEL Classification: I10, I12

Suggested Citation

Evgeniou, Theodoros and Fekom, Mathilde and Ovchinnikov, Anton and Porcher, Raphael and Pouchol, Camille and Vayatis, Nicolas, Pandemic Lock-down, Isolation, and Exit Policies Based on Machine Learning Predictions (January 15, 2022). Production and Operations Management, Special Issue on Managing Pandemics: A POM Perspective, Available at SSRN: https://ssrn.com/abstract= or http://dx.doi.org/10.2139/ssrn.3588401

Theodoros Evgeniou (Contact Author)

INSEAD ( email )

Boulevard de Constance
77305 Fontainebleau Cedex
France

Mathilde Fekom

University of Paris-Saclay ( email )

55 Avenue de Paris
Versailles, 78000
France

Anton Ovchinnikov

Smith School of Business - Queen's University ( email )

143 Union Str. West
Kingston, ON K7L3N6
Canada

INSEAD - Decision Sciences ( email )

United States

Raphael Porcher

Université Côte d'Azur - INSERM ( email )

France

Universite Paris Descartes - Center for Clinical Epidemiology ( email )

France

Camille Pouchol

MAP5, FP2M, CNRS FR 2036, Université de Paris ( email )

45 rue des Saints-Pères
Paris, 75006
France

Nicolas Vayatis

ENS Paris-Saclay ( email )

4 avenue des sciences
Gif-sur-Yvette, 91190
France

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
253
Abstract Views
2,342
rank
165,102
PlumX Metrics