Simple Agent-Based Simulation for Quick Evaluation of System of Systems Architectures

System of Systems (SoS) architectures rely heavily on information exchange to coordinate the independent constituent systems (CS) towards a shared mission goal. However, communication dynamics are often overlooked in the early stage of SoS modelling. This thesis explores how communication and control configurations influence the mission outcomes and individual system behaviour within an SoS. A forest wildfire scenario is modelled using Agent-Based Modelling (ABM) in NetLogo, where multiple firefighting assets such as UAVs, helicopters, water tankers and bulldozers collaborate to suppress the active fire. Three different scenarios of communications were evaluated: a fully decentralised mode where agents act on their own discretion and possess global knowledge of fires, a centralised command where the agents rely on an incident commander for orders and a third scenario, which is similar to the second scenario but adds a delay to regulate the information flow. The obtained results showcase that excessive communication causes command bottlenecks and incomplete missions, thereby affecting the operational outcomes. By throttling communications, the mission completion rates go up, as well as the overall outcomes of the operation. The findings also validate low-fidelity agent-based models as an effective tool to assess SoS architectures and highlight the strategic role of communication in carrying out a successful operation. The work provides a methodological foundation for future architecting of an SoS applicable to any mission-driven operations.