Abstract
In the context of Industry 4.0, the Digital Twin (DT) paradigm is one of the core technologies for establishing Cyber-Physical Production Systems (CPPS). The DT consists in the virtualization of a physical system, where collected data are exploited for performing data processing and/or simulation, whose results are automatically fed back to the manufacturing/assembly system, that will be able to self-optimize itself. A prerequisite for digital twins is to collect realtime data from the physical model and to create a bidirectional data flow between the physical and the digital model, the so-called digital shadow. While the collection of data using sensors, internet of things and cyber-physical systems has been scientifically explored and practical solutions for real-time monitoring have been developed in recent years the step towards digital twins and the exploitation of their potential in industry still poses significant challenges. Developing an easy-to-deploy and simple-to-use DT-based CPPS is still a critical research gap. Integration of DT with secure IoT platforms, the processing of the data using simulation or artificial intelligence / machine learning, and the linking of the data processing results with the higher-level production planning systems and MES systems are a long way off for the current state of implementation of the combination of such technologies. This is where this PhD research comes in, by creating in a first step a DT reference architecture for, thus aiming for a broader applicability of DT in manufacturing industry. Such a reference architecture aims to be reconfigurable, scalable, and (cyber) secure increasing resilience of manufacturing companies. These characteristics make the reference architecture feasible for rapidly evolving manufacturing environment that desire to reach higher productivity by exploiting the potential of digitalization processes.