Modelling Cyber‐Physical Systems Using Data‐driven Patterns

Abstract

The Smart Factory paradigm represents the "fourth industrial revolution" in the field of manufacturing industry, through the implementation of "intelligent systems" consisting of physical systems and software to control and improve manufacturing processes (Zuehlke 2010). These intelligent systems typically include various components, such as sensors for signal acquisition, communication units for data transmission between components, control units for components, control and management units for decision making, and actuators to perform appropriate actions (Lezoche and Panetto 2018). In recent years, the emergence of cyber-physical systems (CPSs) has amplified the ability to sense the world through a network of connected devices using the existing network infrastructure. Cyberphysical system (CPS) aims at embedding computing, communication and controlling capabilities (3C) into physical assets to converge the physical space with the virtual space (Monostori et al. 2016). The combination of intelligent systems and sensing systems forming a large-scale distributed cyber-physical system is a key element in the development of the distributed cyber-physical system. However, they suffer from a lack of modelling techniques that take into account not only their technological parameters but also their high degree of information and functional interrelationships. As the complexity of these systems continues to grow, the challenge of developing intelligent and sensing systems has exceeded the design complexity of their individual components (Lee, Bagheri, and Kao 2015). The main problem in developing intelligent systems is the complexity of integrating and managing these different components, technologies, and objectives across a broad spectrum. In this sense, the concepts defined in the field of Systems Engineering are relevant to the challenge of shared knowledge formalization. It is necessary to define a modelling method that helps to analyse a new form of intelligent systems (smart) and detection in a sustainable perspective. The representation of shared knowledge is a branch of artificial intelligence that studies the way human reasoning occurs and defines symbols or languages. This representation allows the formalisation of knowledge to make it understandable to machines, aligned with reference models. An important prerequisite for the cyber-physical integration is a proper and highly-accurate digital model (Semeraro et al. 2021a). Considering the complexity of digital modelling, this work aims to identify and formalise elements that contribute to the construction of informational and functional models of systems to improve and simplify the modelling of manufacturing processes and products, based on networked components. The idea is to propose a series of modelling patterns aimed at identifying automatically, in the masses of data, invariant behaviours that can be modelled for the emulation of these cyber-physical systems and thus contribute to the digital transformation of industrial production companies. This work a...