The Challenges of Coupling Digital-Twins with Multiple Classes of Faults

In modern industrial contexts, a factory becomes a complex and heterogeneous ecosystem, where many technologies, systems, and workers cooperate. Such a class of systems is named Cyber-Physical Production Systems (CPPSs), since their design requires to merge control, network, and physical aspects. In such a context, it is fundamental to guarantee safe human-machine interactions. Therefore, evaluating and adopting techniques is necessary to ensure functional safety. This article analyzes the challenges of creating digital twins coupled with multiple classes of faults to simulate and verify the system under design. In particular, challenges can be collected under three main categories: modeling, simulation and assessment. Exploiting a language capable of capturing the complexity of such systems is necessary to model CPPSs and support the creation of digital twins. Efficient simulation of CPPSs needs different abstraction techniques and requires to combine discrete and continuous components. Moreover, different classes of faults must be injected into the models to verify the cyber and the physical parts. This would allow assessing the functional safety of each machinery composing the plant.