Digital twin technology, the creation of a digital copy of a device, system, process or person, is fast becoming the norm in innovation, with endless applications in scientific research, manufacturing and healthcare – to name a few. While there is nothing new about considering and manipulating a virtual replica of a physical entity or process, the access in recent years to vastly improved computational power is now enabling this technology to speed up development, reduce production and maintenance costs, and generate safer outcomes.
Speaking at a seminar on digital twin technology hosted by the High Tech Campus Eindhoven, Guido van Gageldonk, Chief Technology Officer at Unit040, described a world of enormous potential: “The technology makes it possible to build visual models that allow all disciplines in a machine development process, say, to see the machine together. Instead of having to work off of a 2D image or line code, multiple teams can see the full picture and can have a good discussion”. This means that, for the first time, so-called agile development is possible with hardware.
Cost savings are a notable benefit. For instance, multiple virtual models can be built, and in parallel; further, the expense of building prototypes is reduced, as fewer are typically needed. Predictive maintenance – being able to model a production system and thus being able to pre-empt system breaks – also saves costs.
Subscribe to IO on Telegram!Subscribe!
Startups can do it, too
Van Gageldonk sees widespread adoption of digital twin technology, and notes a trend towards its use by start-ups. “Smaller companies are now also adopting it; they didn’t have the capital in the past to do it, but the technology is becoming cheaper. In my view it is easier for startups to do this than for bigger companies, because they don’t have the heritage, the legacy, and can do it greenfields.”
In a world where machine learning and artificial intelligence is woven into most aspects of development, digital twin technology is a key element. “It is being used as fuel for machine learning; the digital twin can be used as the system that creates the data. So, you don’t need a larger data set when you start; you can create the data and you can do it cheaply.”
Matthieu Worm, Programme Lead for Autonomous Driving and Vehicle Dynamics: Siemens PLM Software, highlighted the application of digital twin technology to the optimisation of mechatronic systems, using autonomous driving technology as an example.
“The use of digital twin technology is enabling mechatronic system optimisation. More specifically, it is aiding the design of the product, the engineering of the product — mechanics, electronics, software – the design of the manufacturing plant and the process to manufacture the product, and the modelling of performance during the lifetime. It also makes it possible to couple back throughout the lifetime of the product, in order to improve design, manufacturing and use.”
Autonomous driving – where digital twins are especially helpful
In the world of mechatronics, autonomous driving is an application in which the value of digital twin technology is obvious. “The nice thing about this domain is that everyone sees that there is no other way [to the development process] than digitalising the vehicle and the environment,” Worm says. “Perhaps Google and Uber lead the way in bringing that message, but everyone in the automotive industry is adopting the technology.”
One of the challenges for smoother development processes in coming years will be in ensuring more thorough standardisation across processes. Worm refers to the frustration entailed with combining virtual models from different domains. “If you want to build up an entire vehicle, it means bringing together models from different departments using different technologies – bringing different digital twin “children” into a single-vehicle environment. This is highly complex, so the need for standardisation and clearly defined interfaces is very high.”
“Better outcomes, lower costs”
Wim Crooijmans, Innovation Program Manager: Ultrasound and Image Guided Intervention and Therapy, Cardiology at Philips, presented some of the opportunities in healthcare systems for the application of digital twin technology.
He describes his team’s work as “building biophysical models that we can see as a digital copy of functions in the human body, e.g. a heart model, or a model of heart valve replacement procedure. The models are used to plan and execute procedures, with better outcomes and with lower costs”.
A compelling example of this is a model developed by Philips and introduced to the market in 2015, which reportedly is generating an 85% reduction in the time required for a standard cardiac examination.
On the likely path that this technology will take, and its impact on the pace of scientific advancement, Crooijmans is optimistic. “There will be a whole evolution – and perhaps even a revolution. Step-by-step, we will become more accurate in modelling the human body, function and behaviour. The more accurate you are, the better you can predict diagnosis and treatment outcomes.”
Complete modelling of the human will take time
Realism is needed when considering how accurate the modelling of the human body will become, though. “In the coming ten to twenty years we will see many more surprises, with a more complete model. But it will take a long time before we have a full functional representation of the human person,” Crooijmans says.
Chris van Hoof is Senior Director for Connected Health Solutions at imec. His team is applying digital twin technology to building a more complete picture of an individual’s lifestyle, behaviour and environment, which they believe could enable therapists to personalise the treatment of mental health disorders much faster and more effectively.
“The main application of our digital technology R&D is looking at the prevention and interception of disease for which you need a longitudinal – or long-term – recording. We look at which new features you can add or sense, to get a more complete picture of a human in terms of lifestyle, behaviour and varying parameters.”
“Particularly in the field of mental health, the therapy today is very much trial and error. So, a number of these disorders, such as eating disorders, depression or burnout, take a very long time to recovery. The time taken is partly to understand the condition and partly to adopt and adjust the therapy.”
Like Cooijmans, Van Hoof believes there is some way to go before accurate and comprehensive modelling of the human body and behavour is possible. He contrasts this field of study with mechatronic systems, where deep and complete understanding is possible. “Today we can model only a certain number of human body systems. As soon as it is all understood, then of course the sky is the limit and the application is clear and successful.”
Innovation Origins is an independent news platform that has an unconventional revenue model. We are sponsored by companies that support our mission: to spread the story of innovation. Read more.
At Innovation Origins, you can always read our articles for free. We want to keep it that way. Have you enjoyed our articles so much that you want support our mission? Then use the button below: