Adaptive simulation with repast simphony and swift

Abstract

We present a general approach for adaptive ABMS, which integrates Repast Simphony’s distributed batch components and the Swift parallel scripting language. Swift is used to launch Repast Simphony simulations on parallel resources, collect the results from those simulations, and generate further simulations based on an analysis of the results. In order to demonstrate the benefits and capabilities of this approach, we developed a simulated annealing reference workflow and applied it to a modified Repast Simphony “JZombies” demonstration model. The workflow was able to successfully and efficiently find areas of the model parameter space that yielded the desired outcomes, as specified by an objective function. The workflow was run on a high-performance cluster, launching 16 concurrent simulated annealing optimization processes, each executing 100 simulated annealing loops over 16 stochastic model variations – a total of 25,856 adaptive simulation runs, accounting for a 96.3% reduction in the number of simulations that were required compared to a complete enumeration of parameter space. The materials used in the workflow are included as linked external resources to allow replication.