Case study on automated and continuous reliability assessment of software-defined manufacturing based on digital twins.

Traditional production systems are characterized by rare software updates and fixed production lines. Each production unit is designed and programmed for a specific task. Therefore, the reliability assessment is conducted once before the operation, mostly manually, and is based on traditional reliability models, such as event trees, fault trees, or reliability block diagrams. In comparison to traditional production systems, the focus of modern, complex production systems is shifted towards the software part. This is emphasized by the concepts of digital twins and Software-Defined Manufacturing (SDM). These software-intensive and safety-critical systems have more frequent software updates to address higher system flexibility and adjustable production processes. Therefore, SDM systems require a new approach to reliability assessment. Each software update can change the system behavior significantly. This leads to the necessity to reconduct the reliability assessment automatically before each software update. Advanced and hybrid reliability models are the key enabling technology. These models must be automatically generated and synchronized with the available system models and digital twins. Model-to-Model (M2M) transformation methods are another enabling technology. In this paper, we present a case study on automated and continuous reliability assessment of SDM. It shows, that our new method is a suitable candidate to enable the reliability assessment of SDM based on digital twins. The method includes (i) the extension of SysML v2 for reliability assessment, (ii) the automatic generation of hybrid reliability models from the digital twin, and (iii) their reliability assessment with new solvers developed for our OpenPRA framework.