D. Lehner: A Model-Driven Platform for Engineering Holistic Digital Twins, Best Paper Award, Doctoral Symposium, 26th International Conference on Model Driven Engineering Languages and Systems, MODELS 2023, Västeras, Schweden, October 1-6, 2023. Doi: 10.1109/MODELS-C59198.2023.00045
With the combination of software and physical devices into so-called cyber-physical systems (CPSs), Digital Twins (DTs) have emerged to handle the resulting complexity and efficiently connect software to physical devices, the so-called physical twins (PTs). While DTs have gained more and more interest in both industry and academia in recent years, several vendors started to provide so-called DT platforms that offer software tools that promise to make it easier to develop and maintain DTs. When investigating these platforms in more detail, we found that they require the redundant specification of information that is usually already defined in engineering models describing the underlying PT. Additionally, they focus on connecting services to the running PT. Most DT applications however also need a connection to a simulation of the PT, which is currently not supported by the examined DT platforms. As different DT platforms usually each use their own proprietary language and software tooling, it is also currently time-demanding to integrate them with the software services that realize functionality based on these platforms. In the described thesis project, we propose an extended DT platform that solves the mentioned problems by leveraging Model-driven Engineering (MDE) techniques. More precisely, we (i) develop model transformations from existing engineering models to the proprietary DT models used by current DT platforms, (ii) create a DT megamodel that integrates DT models of existing platforms with models representing different endpoints such as PTs or simulations, and generic service descriptions, and (iii) propose a workflow model to define the interactions between different services and DTs, and a method that automates the integration of services and DTs into DT architectures based on this workflow model and the DT megamodel. We aim to evaluate our work by performing case studies on a set of CPSs . In these case studies, we measure the steps required for setting up and maintaining DT architectures for these CPSs, comparing our extended DT platform to existing DT platform support.