System of systems (SoS) engineering is an emerging sub-discipline, within which modelling and simulation is a key area. The failure of many SoS endeavours can be attributed to the inappropriate application of systems engineering processes, including modelling approaches. Selection of suitable modelling approaches is critical to ensuring representative models are created, ensuring reliable SoS analysis and prediction of emergent behaviour is possible. Essential to SoS modelling is the dynamic element, enabling the generation of emergent behaviour, appropriate approaches including; discrete event system specification, Petri nets, agent-based modelling and system dynamics. Bayesian belief networks, Markov models and game theory provide a static probabilistic approach, suitable for strategic, decision and risk models. Enterprise architecture frameworks and modelling languages are typically used for static modelling of SoS. A combination of static and dynamic approaches should be utilised, providing a range of perspectives, giving an insight into the complexity inherent in all SoS.
Keywords: system, system of systems, SoS, system of systems engineering, modelling, simulation, discrete event, agent-based model, neural network, Bayesian, Markov, enterprise architecture, game theory, surrogate model, decision tree