Emergence and Chance in Agent-Based Simulations

Computer-assisted exploratory experiments are used in computational science and engineering as well as computational sociology. Classical (numerical) simulations in science and engineering are mostly based on toy models with very specific hypotheses. However, non-classical simulations have a broader scope: they rely on decentralized information and distributed control thus supporting simulation from-the-bottom-up. Examples include agent-based models (ABMs) and multi-agent systems (MAS). Since emergence is one of the key features of such systems the different variants of emergence in agent-based systems are described. From a system engineer’s perspective the emergent properties, pattern and phenomena in agent-based simulations permit to make creative use of the chance events which may arise during a simulation.