An Agent-Based Model of COVID-19 Diffusion to Plan and Evaluate Intervention Policies

TL;DR

This paper presents an agent-based model simulating COVID-19 spread considering individual behaviors, environment, and intervention strategies, enabling detailed analysis and optimization of policies like vaccine distribution.

Contribution

It introduces a detailed, adaptable agent-based model incorporating behavioral and spatial factors, and employs genetic algorithms for optimizing vaccine allocation strategies.

Findings

The model accurately reproduces multiple waves of COVID-19 in Piedmont.

Simulations show the impact of behavioral and spatial factors on epidemic dynamics.

Optimized vaccine distribution reduces symptomatic cases effectively.

Abstract

A model of interacting agents, following plausible behavioral rules into a world where the Covid-19 epidemic is affecting the actions of everyone. The model works with (i) infected agents categorized as symptomatic or asymptomatic and (ii) the places of contagion specified in a detailed way. The infection transmission is related to three factors: the characteristics of both the infected person and the susceptible one, plus those of the space in which contact occurs. The model includes the structural data of Piedmont, an Italian region, but we can easily calibrate it for other areas. The micro-based structure of the model allows factual, counterfactual, and conditional simulations to investigate both the spontaneous or controlled development of the epidemic. The model is generative of complex epidemic dynamics emerging from the consequences of agents' actions and interactions, with high variability in outcomes and stunning realistic reproduction of the successive contagion waves in the reference region. There is also an inverse generative side of the model, coming from the idea of using genetic algorithms to construct a meta-agent to optimize the vaccine distribution. This agent takes into account groups' characteristics -- by age, fragility, work conditions -- to minimize the number of symptomatic people.