An Architectural Extension for Digital Twin Platforms to Leverage Behavioral ModelsBehaviors

To reduce the work effort for developing and managing Digital Twin (DT)-based services, so-called DT platforms have emerged recently. These DT platforms make it easier to collect, store, and manage data from physical devices, but often they do not provide means for running behavioral models or synchronizing these with runtime data from the physical devices. These aspects are however crucial for developing many DT-based services. As a result, the implementation of a DT that uses behavioral models still requires a lot of implementation effort, which hinders the adoption of DTs for many applications, in particular for small and medium-sized enterprises that often do not have the resources to implement such complex systems. In this paper, we propose an extension to existing DT platforms that simplifies the incorporation of behavioral models. We base this extended platform on requirements derived from two particular use case examples that require behavioral modeling to run their DT services. We discuss the interfaces offered by the extended DT platform, demonstrate its applicability and evaluate the implementation and adaptation effort of our proposed solution in comparison to a baseline implementation. Based on the case studies, the evaluation shows that the extended DT platform reduces the implementation effort by 56% and the effort for adapting DT services across DTs by 39%, compared to existing DT platforms.