Digital Twins Defined
What are digital twins and where did they come from?
What are they?
A digital twin is a digital representation of a physical object, process, place, person, or system that sits at the intersection of connectivity, data and analytics. At a high level, the digital twin concept is comprised of three parts. The physical space, the digital space, and the connection between the two. Digital twins are used in both design, simulation, and operational phases of a product or system lifecycle. Design is done digitally before physical creation occurs, which saves an organization money because they avoid costly, physical mistakes. Digital simulations are performed by feeding test data to see how digital twins react. Think of this as a digital version of a wind tunnel to test the aerodynamics of car. Operationally, data populates the digital twin with the physical counterpart’s current state and behavior. This allows you to observe both the current, and historical state of the physical twin. Further context is derived when additional data from a variety of sources are blended with the digital twin. Software agents compare incoming data from the physical twin with expected state values and KPIs of the digital twin to trigger further analysis and actions. Oftentimes you’ll see hear that digital twins are a 3D CAD model or what you see when wearing AR or VR goggles. Keep in mind that those are actually a view of the digital twin’s data model combined with live data from the physical twin. The view of a digital twin could just as easily be an Excel spreadsheet or exploration through the metaverse.
Where did they come from?
If you remember watching the movie Apollo 13, you saw how Tom Hanks, Kevin Bacon, and Bill Paxton were struggling to get home from the moon in their disabled spacecraft after an oxygen tank exploded. Yes, this happened in real life. You might also remember Gary Sinise troubleshooting the problem in a physical twin of the spacecraft at mission control in Houston. Luckily, through ingenuity and perseverance, the astronauts safely returned to Earth. The learnings from this event gave rise to the idea of creating a high-fidelity, digital replica of spacecraft to make it easier to troubleshoot problems in the future. In the early 1990s, the concept of digital twins were anticipated by David Gelernter’s book, “Mirror Worlds.” In 2002, Dr. Michael Grieves introduced the idea of a “Doubleganger” as part of product lifecycle management (PLM) while he was at the University of Michigan. Dr. Grieves went on to say, “Industry 4.0 is only possible with the digital twin.” In 2010, the name “digital twin” finally stuck when John Vickers of NASA referred to the name in an official roadmap report. Throughout the 2000s, I witnessed the slow rise of digital twins in manufacturing showcased by companies like GE and at events like Hannover Messe.
In 2016, while serving as CTO at Hitachi Insight Group, it occurred to me that digital twin technology should be at the very heart of the industrial IoT platform we were creating. Collaborating with members of the product management and engineering teams, we created an industrial digital twin technology called “Asset Avatars” running within the Lumada platform. The Asset Avatars could model machines, subsystems, assembly lines, and entire factories. We literally brought machines and processes to life in a virtual world to enhance operational efficiency, provide early warning of problems, an ensured uptime of Hitachi assets such as bullet trains and wind turbines. Since then, I’ve been a huge proponent of this powerful technology.
More to Come
Follow along with me as I take you on a deep dive of all the elements that come together to make a digital twin. Click links below to catch up with other articles in the series:
- Digital Twin Prototypes and Models
- Telemetry Properties of a Digital Twin Model
- Static Properties of a Digital Twin Model