Evaluating the reproducibility of multiagent systems

Abstract

A multiagent system (MAS) is a computer based model of multiple interacting intelligent agents within an environment. One of the key aspects of such models is the ability to reproduce the results in different software and hardware. In general, the simulation of these models are carried out under floating point standards. It is well known that by limitation of computer arithmetic, the results may be different when the combination of hardware and software is also different. In such case, the evaluation of the reproducibility of MAS is harder than deterministic models, as it is usually seen as stochastic process. To overcome this problem, this paper presents a Kolmogorov-Smirnov test to verify if two simulations of a MAS keep the same statistical properties. Computational simulations illustrate the efficience of the technique as a tool to verify the reproducibility of a MAS.