Why you need a digital twin

Digital twins are multiplying in the IT sector. Defined by the Defense Department as a simulated model of a physical process, product or service that "uses sensor information, and input data to mirror and predict activities/performance over the life of its corresponding physical twin."  They can help government agencies operate more efficiently and at lower costs.

The idea of the digital twin can be traced to NASA’s need to operate, maintain and repair systems in space from Earth.

“The ultimate vision for the digital twin is to create, test and build our equipment in a virtual environment,” John Vickers, NASA’s leading manufacturing expert and manager of NASA’s National Center for Advanced Manufacturing, told Forbes. “Only when we get it to where it performs to our requirements do we physically manufacture it. We then want that physical build to tie back to its digital twin through sensors so that the digital twin contains all the information that we could have by inspecting the physical build.”

The Air Force uses digital twins to track fatigue in components of individual aircraft, improving lifecycle management. It integrates existing technical data, material properties, flight data, maintenance and inspection reports as well as probabilistic analysis to create an airframe digital twin that provides data on tailoring maintenance by aircraft tail number.

Most current examples of digital twins at U.S. government agencies tend to be on a relatively small scale: smart parking initiatives, connected de-icing machines at airports and smart street lights. Abroad, Boston uses a limited version of a digital twin on construction projects to evaluate how much shade or shadow proposed buildings will cast throughout the year. Singapore has a 3D digital twin called Virtual Singapore, a $73 million living replica of the city that officials use for planning and communicating with residents.

The two main reasons to use digital twins relate to asset management and maintenance and business needs, said Alfonso Velosa, an analyst at Gartner’s Internet of Things Group. For instance, a building manager with a digital twin of the structure could see immediately when an elevator is out of order rather than having to wait for someone to report the problem.

As cities are evolving their approach to public parking, they are replacing in-ground sensors with smart cameras that can cover a large area, recognize a car and clock how long it’s been in a parking lot – data that replicates part of the real-world environment. Now, instead of purchasing and installing multiple sensors, cities can get the same information from a few cameras, Velosa said.

“We are seeing folks explore monetization and doing R&D and looking at new business models, but I’d say 90 percent of my conversations right now -- it doesn’t matter who -- are about maintenance and reliability and about business process and asset optimization,” he said.

The global market for digital twins is expected to grow by almost 38 percent annually, reaching $15.7 billion by 2023, according to MarketsandMarkets research. Forty-eight percent of organizations that are implementing IoT said they are using or plan to use digital twins this year, a March Gartner report found.

Although digital twins are still in the nascent stage, Velosa said, his company’s August Emerging Tech Hype Cycle report puts them midway between discovery and adoption, with widespread adoption likely within five to 10 years.

One driver of this growth is the proliferation of enabling technologies, such as IoT and machine learning, deployments of which are each expected to almost double by 2020, according to a Deloitte report. “As the technologies that enable digital twins become more cost-effective, enterprises today are finding more ways to benefit from these connected digital avatars,” it states.

But digital twins don’t have to be complex, Velosa said: “This is not rocket science. … A digital twin can be a basic graph, it can be just a few lines in the spreadsheet because I might just need to know, ‘Is it on? Is it off?’”

Right now, the main challenges agencies face in creating digital twins include a lack of clear standards for implementing them, a need to train people to use them and a plan for governance, Velosa said. “You could have a model, for example, of a pump and that model belongs to the [original equipment manufacturer] of the pump, but if that pump is operating at your local power plant, the data from that belongs to the power company,” he said. “Clearly, the OEM builds the pump’s digital twin, but who builds the digital twin of the power plant?”

Another stumbling block is marrying discrete digital twins into a bigger composite, such as taking digital twins used by a utility company and a smart parking application and fitting them into an overall digital twin of the city, Velosa said.

These challenges are worth tackling though, he added.

“There’s a reason to do this, and in effect, digital twins are a management model that can be used to improve enterprise decisions,” Velosa said. “This is sort of a precursor to us continuing to build this digital universe we’re headed toward.”