Life Cycle-Spanning Experimentable Digital Twins

Abstract

Digital Twins (DT) revolutionize our view on systems—and this from completely different perspectives. During development, the twin type is developed and tested via DT instances, reducing the need for early prototypes of the corresponding Real Twin. In the context of networking, e.g., on the Internet of Things (IoT), DTs are networked and thus indirectly the corresponding real machines. In the context of user interfaces needed to supervise and command a machine, the worker interacts with the DT, which then performs appropriate actions on the real machine. In the context of the development of Artificial Intelligence (AI) algorithms, DTs generate the necessary training data in the virtual world in a short time without endangering people or machines. The same applies to the training of drivers or operators, who can safely learn how to operate a machine via the DT. In Semantic World Modeling (Sondermann 2018), DTs that represent objects in the environment are created as a basis for subsequent planning activities. Further examples of the use of DT can be found in areas such as predictive maintenance, model-based systems engineering, model-based control, or virtual commissioning (VDI 2016).