Machine Learning Model for System-of-Systems Interoperability

The article discusses the role of digitalization and the Internet of Things (IoT) in today's world, with a focus on their application in industrial automation as part of the Industry4.0 concept. A case is made for the use of IoT to improve monitoring, increase production flexibility and save resources. Particular attention is given to systems of systems (SoS), which consist of autonomous and interacting components. Five key characteristics of SoS are examined: autonomy, ownership, connectivity, diversity, and emergence. The paper also analyzes service-oriented architectures (SOAs), such as RAMI4.0 and OPC-UA, which are used to integrate IoT and SoS. The importance of developing interfaces and service contracts to enable interoperability between system components is discussed. A separate section is devoted to cyber-physical systems (CPS), which combine physical and computational elements to create more complex and interconnected industrial systems. The paper discusses the different classes of interoperability: informational, semantic, dynamic, and operational, and their role in the design and operation of SoS. It concludes by proposing the application of network control principles SPCFA to SoS system management, which contributes to improved monitoring and control in industrial automation.