A fuzzy cellular automata for SIR compartmental models

Abstract

We propose a representation of the dynamics of epidemics through a compartmental SIR (Susceptible - Infected - Recovered) model, with the combined use of geo-referenced cellular automata and fuzzy systems. In this model, each cell does not correspond to an individual, but to groups of individuals inhabiting the physical space corresponding to the cell. The temporal evolution of the transmission consider is modeled by changes in the size of groups of individuals in each category (susceptible, infected and recovered). We applied our model on the spread of dengue in a region in Southeast Brazil. The application shows that the proposed model, using only a small set of simple fuzzy rules, is able to represent qualitatively the behavior of an epidemiological SIR mode, a rather complex problem. © Springer International Publishing 2013.