Opening pandora's box: Some insight into the inner workings of an Agent-Based Simulation environment

Agent-Based Simulation (ABS) environments are somewhat of a black box to many modelers in Social Simulation or Economics and their inner workings are often only understood by the computer scientists who developed them. We intend to shed some light into the inner workings of such systems. For this purpose we have developed our own simple ABS environment in C++ using hierarchical state machines. In this paper we provide insight into the design of our ABS environment and then test the performance of it by comparing it to that of an "off the shelf" commercial package. While some programming knowledge is required to understand the paper in all its depth we believe that non programming experts will also benefit from this paper as it provides an insight into the underlying mechanisms operating within an ABS using graphical representations and explanations that avoid heavy technical jargon.