DEVS re-implementation of an agent-based valley fever model

Discrete Event Simulation is a natural way to simulate models that are characterized by heterogeneity in their spatio-temporal behavior. This paper describes a valley fever spore-propagation model that describes the time dependent spatial distribution of spores in southern Arizona. Two implementations of this model are examined and their execution times are compared. A discrete event model exploits the fact that only a few areas are favorable for the growth of the fungus at any given time. This allows it to schedule events only at the areas that are favorable. In contrast, the existing discrete time model cannot take advantage of the few areas that are favorable.