The art of iterating: Update-strategies in agent-based simulation

Abstract

When developing a model for an Agent-Based Simulation (ABS) it is very important to select the update-strategy which reflects the semantics of the model as simulation results can vary vastly across different update-strategies. This awareness, we claim, is still underdeveloped in the majority of the field of ABS. In this paper we propose a new terminology to classify update strategies and then identify different strategies using this terminology. This will allow implementers and researchers in this field to use a general terminology, removing ambiguities when discussing ABS and their models. We will give results of simulating a discrete and a continuous game using our update-strategies and show that in the case of the discrete game only one specific strategy seems to be able to produce its emergent patterns, whereas the pattern of the continuous game seems to be robust under varying update-strategies.