Towards Linked Data for Ecosystems of Digital Twins

Due to either the inherent complexity of the domain or the evolving nature of systems, we can envision solutions that digitalize assets in a complex domain using an ecosystem of distributed Digital Twins instead of a single monolithic one. To effectively tackle interoperability in such ecosystems, this paper advocates for the introduction of a representation based on Semantic Web technologies enabling the discovery of both Digital Twin structure - i.e. the static information about the asset model and offered services - and state - i.e. the data and metrics collected at runtime - to support the management of ecosystems and the creation of application mashups. A review of the state of the art suggests that currently investigated ways to describe a Digital Twin are not sufficient to achieve this objective. A proposal of key requirements for a Digital Twin representation is outlined leading to the proposal of a core ontology and a Linked Data approach for state management.