A cellular automata to model epidemics

Abstract

Compartmental models are very popular in epidemiology, they provide excellent results when the populations satisfy certain hypotheses as large population size or population homogeneity, the complexity of this models is low making their analysis intuitive. In the other hand, they ignore important factors inherent to the problem, such as the nature of contacts between individuals and population heterogeneity. Cellular automata models are adequate to describe natural systems consisting of a massive collection of simple objects. They represent the global system behavior as a colection of simpler objects or cells. In this paper we propouse a cellular automata model to study the time evolution of a heterogeneous population through the various stages of disease resulting from the individuals interactions (epidemic). We validate the model with real data of flu that hit Geneva (Switzerland) in 1918 and then we will test the model under different assumptions discussing the result that each has on the disease dynamics.