Beginner's Guide To Digital Twin Technology

We look at how it is changing manufacturing


Digital twin technology has seen a dramatic increase in hype over the last two years, even making Gartner's 2017 Hype Cycle for emerging technology. But what is it, and should you be looking to implement the technology?

A digital twin is essentially a computerized mirror of a physical asset and/or process, a virtual replica that relies on real-time data to mimic any changes that occur throughout its lifecycle. The idea was conceived by Michael Grieves at the University of Michigan in 2002, where it was referred to as the Mirrored Spaces Model in the first executive PLM courses. However, until recently, it was only available to organizations with the resources to collect, store, and analyze the massive amount of data that underpins it. It was also restricted by limitations in digital technology capabilities. Today though, advances and improved cost-effectiveness in machine learning, IoT, and cloud connectivity mean that companies in a variety of industries are empowered to implement digital twin technology.

In order to create a digital twin, the physical asset is built with one or more sensors that collect real-time data and operational status. This is sent via a cloud-based system before being analyzed, using machine learning algorithms in cases where there are more complicated variables and a harder-to-discover linear relation. The changes identified are then replicated in the twin. The twin can vary widely in complexity. Alfonso Velosa, research vice president at Gartner, notes that, 'The complexity of digital twins will vary based on the use case, the vertical industry and the business objective. In some cases, we will have simple, functional digital twins that are based on clearly defined functional or technical parameters. In other cases, they may require physics-based high-fidelity digital twins. In still other cases, there are compound systems composed of other digital twins that need to be integrated.'

A digital twin has many benefits. Indeed, IDC predicts that this year will see companies who invest in digital twin technology improve cycle times of critical processes by 30%, while Gartner believes adopters see a 10% improvement in effectiveness. The German Association for Information Technology, Telecommunications and New Media (BITKOM), meanwhile, estimates that every digital twin in the manufacturing industry will have an economic potential of more than €78 billion by 2025. Essentially, digital twin technology is useful in that it enables you to run simulations of a product's performance, with the impact of different variables tested from the design stage onwards. This enables you to predict potential problems, overheating, and the like, in the physical machine before they become reality. You can then refine or correct processes before any issues arise, thereby reducing downtime, overhead expenditures, and ultimately extending the lifespan of physical assets. Furthermore, endless design iterations can be tested in the virtual world without having to stop the production line to see how they can be integrated.

Digital twin technology was first used in practice by NASA, who needed to mend, update, and monitor machines in outer space, where it is often impossible to be physically present. It proved vital during the failure of the Apollo 13 mission, for one, with engineers and astronauts able to use digital twins to establish what was going wrong and fix issues remotely. It has proved similarly successful at a number of major businesses. At the Siemens Amberg plant, which produces industrial computer-control systems, the technology has been integral as they scaled production to 15 million units a year without expanding the building or hiring more people. They created a virtual replica of the factory, identical to the physical one. The Economist notes that, 'This digital twin is identical in every respect and is used to design the control units, test them, simulate how to make them and program production machines. Once everything is humming along nicely, the digital twin hands over to the physical factory to begin making things for real.' And the benefits they've seen have been tremendous, helping to slash factory's defect rate to almost zero, with 99.9988% of units requiring no adjustment.

Another company to have invested heavily in digital twin technology is GE. The energy giant has implemented over 500,000 digital twins, which it uses to inform the configuration of each wind turbine before they are built. Ganesh Bell, chief digital officer and general manager of Software & Analytics at GE Power & Water, says that, 'For every physical asset in the world, we have a virtual copy running in the cloud that gets richer with every second of operational data.' GE’s 'digital wind farm' has enabled them to push for 20% gains in efficiency through analysis of the data from each turbine that is fed to its virtual equivalent.

At the moment, these companies remain in the minority, but growth is expected. Gartner predicts that by 2021, half of large industrial companies will use digital twins, while new market research 'Digital Twin Market by End User (Aerospace & Defense, Automotive & Transportation, Home & Commercial, Electronics & Electricals/Machine Manufacturing, Energy & Utilities, Healthcare, Retail & Consumer Goods), and Geography - Forecast to 2023', put growth in the digital twin market at 37.87% CAGR), up to $15.66 Billion by 2023. There are still challenges that needs to be overcome though. Gartner analyst Marc Halpern warns that 'It will take longer and will be more resource-consuming than anyone can imagine to get these solutions in place.' However, while the cost and time of bringing in digital twins may still remain prohibitive for many, as the technology improves, it is likely that we will be seeing far more of digital twins in years to come.  


Read next:

Why Blockchain Hype Must End