October 22, 2019 - Imagine the cost savings that could come from having a perfect digital copy of an assembly line product – a replica that isn’t just about dimensions and assembly details, but also reflects materials, lifetime maintenance data, and more.
Conceivably, it could be used to ferret out savings in labor, materials, and manufacturing costs; it could help predict failure modes and MTBF in the real world; and help hone improvements for future iterations of the product.
Built entirely from numbers, this digital copy could be used in simulations, analysis and analytics, changing the very way we think about product design, manufacturing, and improvement processes.
A hot topic
It’s already got a name: A digital twin. And it’s a hot topic right now, making Gartner’s list of top tech trends two years in a row.
In 2017, Gartner predicted that within three to five years, hundreds of millions of things will have their own digital twins. In 2018, Gartner doubled down on those estimates by saying – in the context of the Internet of Things – that “with an estimated 21 billion connected sensors and endpoints by 2020, digital twins will exist for billions of things in the near future.”
Interestingly, digital twins don’t even need to be things. In the same 2018 report, Gartner suggested that eventually, “digital representations of virtually every aspect of our world will be connected dynamically with their real-world counterparts and with one another and infused with AI-based capabilities to enable advanced simulation, operation and analysis.”
And while there are some ambitious ideas for a future filled with digital twins that mimic people (and their biometric and medical data) and even entire cities for every aspect of digital city planning, already some industries are creating digital twins not just for products but for their manufacturing processes as well.
The use cases
There are myriad use cases for digital twins. Virtually anywhere you can find a place for the IoT, there’s likely an application for digital twins.
Some of the most obvious applications for digital twins can be found in manufacturing, where digital twins of complex products can play a role at every stage of the product lifecycle.
In the design and development phase, the digital twin (combined with other tools, like machine learning AI), can find the most efficient way to manufacture the product, minimizing materials and workflow steps.
Later, after sale, the digital twin can receive sensor data about performance of its real-world siblings to simulate and evaluate characteristics like wear and tear.
That, of course, helps engineers deliver just in time predictive maintenance and engineer improvements for future versions.
Improving customer experiences
Meanwhile, in the retail industry, digital twins can help improve the customer shopping experience.
With increasingly accurate models of consumers, manufacturers and retailers can make it easy to offer virtual sales experiences, in which customers have confidence that products will look good and fit well without having to try it on in real life.
Likewise, the healthcare industry is slowly but surely embracing digital twins to enable personalized delivery of medical care.
And the automotive industry can not only benefit from many of the same use cases as manufacturing, but also can create virtual models of connected cars. The result: smarter self-driving vehicles that learn from huge volumes of real-world driving behaviors.
Finding valuable roles in many industries
Digital twins would not have been possible even a decade ago.
They’re enabled by a few key technologies. First and foremost is the Internet of Things – all the sensors which instrument the products, assembly lines, people, and processes that make it possible to monitor, simulate, and analyze the behavior of real systems in the virtual world.
But just as important is the rise of simulations tools that can design and visualize these objects and systems in sufficient detail.
And then there’s the AI tools, like machine learning, which is a uniquely 21st Century field of study. Without the advances in AI we’ve experienced in the last decade, it wouldn’t be possible for software to make meaningful predictions and recommendations with the digital twins.
Will digital twins soon number, as Gartner predicts, in the billions? These kinds of tea leaf readings are always a bit murky, but one thing is certain: there is no question that digital twins are finding valuables roles throughout many common industries, and their fortunes look to only improve as the core tech behind it – IoT, design software, and AI – gets faster, better, and more pervasive.