Spreading of infections on random graphs: A percolation-type model for COVID-19

TL;DR

This paper presents a percolation-based epidemic spreading model on networks, extending the SIR framework to incorporate lockdown effects, and demonstrates its effectiveness in fitting COVID-19 data with notable differences from traditional models.

Contribution

The paper introduces a novel percolation-inspired epidemic model that extends the SIR model to better incorporate lockdown effects and provides detailed simulations and empirical data comparison.

Findings

Both models fit COVID-19 data well

Differences between models highlight alternative approaches' usefulness

Percolation model offers new insights into epidemic spread dynamics

Abstract

We introduce an epidemic spreading model on a network using concepts from percolation theory. The model is motivated by discussing the standard SIR model, with extensions to describe effects of lockdowns within a population. The underlying ideas and behavior of the lattice model, implemented using the same lockdown scheme as for the SIR scheme, are discussed in detail and illustrated with extensive simulations. A comparison between both models is presented for the case of COVID-19 data from the USA. Both fits to the empirical data are very good, but some differences emerge between the two approaches which indicate the usefulness of having an alternative approach to the widespread SIR model.