The Evolution of the Digital Twin

A visionary product concept brings big changes for the future

Ein Beitrag von: Michael Grieves, Florida Institute of Technology Center for Advanced Manufacturing and Innovative Design (CAMID)

The Digital Twin is a product manufacturing concept that is receiving much current attention. While the concept originated over a decade ago, the advances in information technology have now made the Digital Twin feasible to implement. The attractiveness of the Digital Twin is that it promises to provide value for both manufacturers and customers throughout a product’s creation and life. There are several types of the Digital Twin. Digital Twins will increase in intelligence, which will be critical as Artificial Intelligence (AI) moves into products.

The Digital Twin model made its public debut in December 2002 at the University of Michigan. It was part of a presentation at the formation meeting of the proposed Product Lifecycle Management Center. The attendees included executives from the Big Three Detroit automotive companies (General Motors, Ford, and Chrysler), Tier 1 automotive suppliers, and Product Lifecycle Management (PLM) software providers. The model did not have a name at that time. The slide was simply entitled, „Conceptual Ideal for PLM“. However, the label was meant to convey that PLM was all about capturing, maintaining, and using information about a product. It was the idea that all the information about a product could be mirrored with information in PLM.
The idea that there is an informational “twin” of physical products, had been an area that was explored during a doctoral program at Case Western Reserve University. Informational Twins have existed as long as humans have had things that they made or what we call “products.” Informational twins first existed as ephemeral thoughts in people‘s minds and then progressed to a tangible permanence as people began to capture their thoughts on paper and physical scale models. The rise of computers in the last half of the 20th century and their exponential advancement as described by Moore‘s Law meant that we were rapidly reaching the point where a Digital Twin that could fully mirror a physical product was not too far off.
The value of the Digital Twin (and informational twins) is that information is a replacement for wasted physical resources: time, energy, and material. While physical resources have to be expended in activities in order to accomplish desired results in our physical world, knowing exactly what activities will produce those results can dramatically reduce physical resource usage. Using digital bits that get cheaper every day is usually far more efficient and effective than using physical atoms that get more expensive every day.
Admittedly the Digital Twin was ahead of its time when it was introduced. The Digital Twin’s real usefulness also required that PLM move from the create phase in engineering into the build phase of manufacturing and the usage/support phase when the product is placed into service. By around 2015 both the technology of modeling and simulation, and the usage of the digital tools throughout the product lifecycle made the Digital Twin a reality.
In the past year, there has been acceleration of the usage of the Digital Twin as both manufacturers and users have recognized its value in mirroring a product’s creation, manufacturing, and performance. Gartner, Inc. a leading information technology research firm, named the Digital Twin as one of the Top 10 Strategic Technology Trends for 2018.

The Digital Twin and Its Types

The Digital Twin model is made up of three parts, the physical product (which can also be thought of as the Physical Twin), the Digital Twin, and the communications of data and information between the two. The Digital Twin model requires all three of these components in order to be useful. What would be helpful are some definitions to rely on when referring to the Digital Twin itself and its different manifestations in different phases of the product’s lifecycle. The following definitions are proposed:

→ Digital Twin (DT): the Digital Twin is a set of virtual information constructs that fully describes a potential or actual physical manufactured product from the micro atomic level to the macro geometrical level. At its optimum, any information that could be obtained from inspecting a physical manufactured product can be obtained from its Digital Twin. Digital Twins are of three types: Digital Twin Prototype (DTP), Digital Twin Instance (DTI), and Digital Twin Aggregate (DTA).

→ Digital Twin Prototype (DTP): this type of Digital Twin describes the prototypical physical product. It contains the informational sets necessary to describe and produce a physical version that duplicates or twins the virtual version. While the Digital Twin is thought to be the “twin” of an existing product, the reality is that the Digital Twin iscreated before there is a physical product. The ideal is that we would like to develop the Digital Twin, test it to ensure that it produces the behaviors we desire, manufacture it digitally to make it as lean as possible, then use and support it digitally, simulating it throughout its life cycle. Only when we are convinced that the product design meets all our requirements, we would then make the physical product. Again, if we are seeking the ideal, we would print it using Additive or 3D Manufacturing! Clearly the complexity of the product drives the amount and types of data and information needed. A paper clip doesn’t need a DTP. A 21st century fighter jet needs a substantial and complex DTP.

→ Digital Twin Instance (DTI): this type of Digital Twin describes a specific corresponding physical product that an individual Digital Twin remains linked to throughout the life of that physical product. Depending on the use cases required for it, this type of Digital Twin may contain many different types of information: geometrical information, service information, part changes, etc.

→ Digital Twin Aggregate (DTA): this type of Digital Twin is the aggregation of all the DTIs. Unlike the DTI, the DTA may not be an independent data structure. It may be a computing construct that has access to all DTIs and queries them either ad-hoc or proactively.

What are the major uses of Digital Twins?

The two major uses are predictive and interrogative. Predictive – the Digital Twin would be used for predicting future behavior and performance of the physical product. At the Prototype phase, the prediction would be of the behavior of the designed product. In the Instance phase, the prediction would be of potential component failures of an individual product: an airplane, pacemaker, turbine.
Interrogative – Digital Twin Instances could be interrogated for the current and past histories. Irrespective of where their physical counterpart reside in the world, individual instances could be interrogated for their current system state: turbine blade speed, geographical location, structure stress, or any other characteristic that was instrumented. Multiple instances of products would provide data that would be correlated for predicting future states. For example, correlating component sensor readings with subsequent failures of that component in a population of products would result in an alert of possible component failure being generated when that sensor pattern was reported.

Intelligent Digital Twin (IDT)

The initial discussions of the Digital Twin focused on it as being a passive repository of the physical product information that was able to be interrogated. However, the idea of an active or intelligent Digital Twin was always a part of the concept from the beginning. One of the key characteristics of PLM that was articulated initially was called “cued availability.” While not a particularly elegant term, cued availability meant that the Digital Twin would present appropriate information based on the contextual cues of the product status in the environment that it was operating in. With the current rapid advancements in Artificial Intelligence (AI), combining the DigitalTwin with AI leads to the Intelligent Digital Twin.
One of the major opportunities for an IDT is to do Front Running Simulation (FRS). What FRS does is to take the current state of the product and, based on its modeling and simulation of behavior, predict the product’s state into the future, in minutes, hours, or days. This would be useful in predicting product degradation or failure, bottlenecks in factories that would occur, or actions with the operation of equipment that could lead to unsafe or catastrophic conditions. With the rise of IoT, there is much discussion about putting AI in the equipment itself to modify equipment behavior and responses, i.e., “learn”, as time goes on. This sounds reasonable until one realizes that there is a real possibility of unintended behavior cascading out of control in a disastrous fashion. Having an IDT to provide overwatch and FRS capabilities, intervening and bringing a product back under control needs to be a precondition before putting AI in products.