Report of TWG Digital Twins: Landscape of Digital Twins

Digital Twins represent one of the most novel and promising concepts within the general trend of digital transformation. They are a distinctly 21st century invention. The term is attributed to Prof. Michael Grieves of the University of Michigan in 2002, and the approach was adopted in particular by NASA around 2010. The concept has proved very powerful and has now been universally used across domains, and yet we have only scratched the surface of what they can or should do. While the current generation of digital twins are replicas of existing entities in the physical world, future twins will help design or invent things that have not yet been realized, bridging the gap to virtual reality. As befits such a new approach, the definition of digital twins is not quite stable yet, and standards are only barely emergent. While Wikipedia defines a digital twin as “a virtual representation that serves as the real-time digital counterpart of a physical object or process,” various organizations have attempted to add more precision to the definition. The Digital Twin Consortium’s definition is “a virtual representation of real-world entities and processes, synchronized at a specified frequency and fidelity.” This is a more operational definition – one that insists on the nature of the link between the two twins – the digital one and the physical one. It has the merit of explaining that a digital twin is not just a simulation model – there are data and/or commands that get sent by one twin to the other. Whenever a new technology emerges, standards are often at first controversial. They are seen as premature – potentially interfering with the exploratory nature of the new idea, and limiting the freedom of companies, especially technology suppliers, to invent and patent new intellectual property. Over time, the ideas and mechanisms that contribute to the technology tend to stratify: there is a layer of foundational components whose standardization increasingly appears useful to all parties, because such standards reduce the “friction” in the development of useful systems. For example, a multiplicity of data formats impedes the ingestion, combination, and processing of information. Conversely, the application level remains the one at which competition creatively rages and standardization continues to be considered harmful. Digital twins have not even reached the point at which specific, normative standard specifications have emerged for components of the technology prope