Simulation of influenza epidemics with a hybrid model - combining cellular automata and agent based features

To understand and predict epidemic patterns ODEs and PDEs have been used since the beginning of the last century. But these approaches have a quite relevant shortcoming. Trying to model a multiply heterogeneous population (e.g. with individual characteristics, varying population densities) increases complexity beyond limits. To bring individual effects into epidemic models a new approach is necessary. Agent-based (AB) models as well as cellular automata (CA) represent tools which allow incorporating such influences. In this paper we shall present a hybrid model that combines the flexibility of an AB-framework with the computational efficiency of CAs. We will also look at the potential benefit of such a structure by taking a look at first (academic) results.