Data-driven Optimization Model for Global Covid-19 Intervention Plans

TL;DR

This paper introduces a data-driven integer programming model to optimize COVID-19 intervention strategies, balancing infection reduction and economic impact, based on historical data analysis and empirical evaluation.

Contribution

It presents a novel optimization framework for COVID-19 interventions that integrates data-driven impact estimation with multi-objective decision-making.

Findings

Model effectively balances health and economic outcomes.

Empirical results outperform simple heuristics.

Provides visual tools for policy analysis.

Abstract

In the wake of COVID-19, every government huddles to find the best interventions that will reduce the number of infection cases while minimizing the economic impact. However, with many intervention policies available, how should one decide which policy is the best course of action? In this work, we describe an integer programming approach to prescribe intervention plans that optimizes for both the minimal number of daily new cases and economic impact. We present a method to estimate the impact of intervention plans on the number of cases based on historical data. Finally, we demonstrate visualizations and summaries of our empirical analyses on the performance of our model with varying parameters compared to two sets of heuristics.