If manufacturing was a series of blockbuster movies, the latest theater marquee might read, “Coming soon to a supply chain near you: The Digital Thread.” And John Vickers would be one of the first in line to buy tickets. The principal technologist for the Space Technology Mission Directorate at NASA Headquarters in Washington, D.C., he has an agency-wide responsibility for its advanced manufacturing activities—and plays a key role in helping to guide the organization forward in an increasingly digital world.
Vickers rarely uses the phrase “digital thread,” however. Instead, he favors the more comprehensive phrase “digital twin,” a term that he said lies at the heart of any digital transformation.
“The digital thread is but a small piece of the digital transformation currently underway at NASA and elsewhere throughout the manufacturing community,” he said. “Much of this revolves around the use of the digital twin, or more broadly, a concept that we refer to as ‘model-based everything’.”
As he explained in a recent presentation to the Cambridge Group, the digital twin dramatically enhances NASA’s mission impact by “engaging digital convergence, reinventing mission and mission support processes, products, and capabilities.” Here again, though, his definition of the digital twin and other industry standard terminology is not always in lockstep with that of his colleagues.
“It seems that everybody tries to define the digital twin in very prescriptive terms, but I refuse to do that,” Vickers said. “For instance, I propose that it does not require a physical asset, a viewpoint that some of my friends at the AIAA [American Institute of Aeronautics and Astronautics]—who recently published a paper on digital twins—disagree with. Nor is it synonymous with other technologies, such as MBSE [model-based systems engineering] or, as I already mentioned, the digital thread, even though it contains elements of each.”
As designed, as built, as operated
The digital twin is an interdisciplinary approach, he explained, one that allows manufacturers to analyze, synthesize and harmonize links between disciplines into a coordinated and coherent whole.
It is “collaborative, predictive, descriptive, investigative, cognitive, and corrective.”
And while Vickers’ version of the digital twin is indeed model-based, it’s that first part—collaborative—that prevents users from “throwing it over the wall” as with traditional multidisciplinary models, which tend to keep information within their own siloed environments.
This means that the digital twin used today during the design and manufacturing stages will one day drive the entire business enterprise. This includes the marketing, management, production, and finance groups, and ultimately, the product’s end-users, which in NASA’s case, might be taking those products to the moon or beyond.
Birth of the twin
Vickers pointed out that he and NASA consultant Michael Grieves—now the chief scientist for advanced manufacturing at the Florida Institute of Technology—coined the phrase “digital twin” in 2010. And while it’s not yet a teenager, the digital twin of that era has evolved greatly over the past decade.
For instance, there are far more advanced simulation and analytics systems available today, as well as machine learning and artificial intelligence.
All now play significant roles in any digital transformation initiative.
So do augmented, hybrid and virtual reality tools. These help humans to visualize and test virtual products, then teach them how to operate their physical versions once they’ve been deployed.
And of course, there’s additive manufacturing, which for NASA and many others is a key enabler of faster, more cost-effective product design.
Don Kinard is a senior fellow for Lockheed Martin’s aeronautics production operations in Fort Worth, Texas. He also has decades of experience with model-based engineering, a trend that began in earnest during the early days of the F-35 joint strike fighter program.
“The F-35 signified the beginning of our digital transformation,” he said.
“Unlike its predecessor the F-22, which was still paper-based, it was the first entirely digital engineering aircraft program. We had solid models for everything.”
That was in 2004. Since then, digitization has brought countless benefits to Lockheed Martin.
Aside from the obvious ones, such as more efficient design and engineering processes, it has also enabled significant improvements on the manufacturing floor.
This includes automated drilling and fastener installation, improved machining processes, robotic spraying of protective coatings, computer-controlled laser-cutting of tubes and, more recently, non-contact metrology—all driven by digital data.
Non-contact metrology is significant in many ways, Kinard noted. By comparing 3D solid models to structured light scans of aircraft structures and subassemblies, manufacturers find it both faster and easier to answer any questions about as-designed vs. as-built.
“It’s our job as the technology group to identify what the production floor needs and where opportunities for automation exist, then go figure out how to implement them cost-effectively and with a solid return on investment,” he said. “In many cases, the solution is a digital one.”
None of this is new, he added. What has changed are the wealth of tools available to manufacturers today, whether that is the structured and laser light scanners just mentioned or the advanced-analysis software tools and systems used to analyze aircraft design.
“A few years into the F-35 program, I could see very clearly how much of a difference model-based engineering makes, starting with the initial design of the aircraft all the way through to how we support it in the field.”
Enter the Matrix
It is also affecting how aircraft parts and materials are sourced.
Kinard pointed to the works of Will Roper, the assistant secretary of the Air Force for acquisition, technology and logistics. In his “Bending the spoon” papers, Roper wrote, “Though our Cold War process does produce world-leading military systems, it is escalating timelines and costs are unsustainable byproducts. The stark contrast with commercial industry puts our military at the ‘wonderless’ end of the rabbit hole.”
According to Roper, the way out of this rabbit hole is through digital engineering, a set of technologies that has led to the USAF’s “e-Series” designation for aircraft, satellites, and weapon systems wholly designed and manufactured on a digital foundation.
The first member of this rapidly growing club? The eT-7A Red Hawk, a jet trainer designed and built in just 36 months—and named in honor of the Tuskegee Airmen.
“The ability to develop virtual prototypes early on in the development phase lowers the risk for production, since it lets us know whether the design will meet the customer requirements before we actually start cutting metal and laying down composites, let alone spending years on flight and structural tests” Kinard said. “So that is really the emphasis today, most of which revolves around simulation modeling. Our world is going to change dramatically over the next decade or so as these technologies grow more sophisticated and the fidelity of our 3D models increases.”
Paul Oldroyd, who serves as a technical fellow and principal technical resource for manufacturing and process development at Bell (a division of Textron), agreed—but with one caveat: Even considering its remarkable successes and embrace of the digital transformation, the industry still has a ways to go.
“The word ‘transformation’ infers a dynamic environment, which means we must continue to move forward,” he said. “Still, we have all made progress toward a fully digital architecture.”
You ain’t seen nothin’ yet
Bell has certainly come a long way since the V-22, its first FBW (fly by wire) aircraft, he explained.
Additionally, the 525 Relentless will be a fully FBW commercial rotorcraft. The Joint Multi-Role (JMR) aircraft including FLRAA (Future Long Range Assault Aircraft) and FARA (Future Attack Reconnaissance Aircraft) are being developed using the digital twin—and have realized significant benefit from incorporation of the digital thread.
“As a discrete example of the benefit, the JMR Valor V-280 nacelle hydraulic system realized a 90-percent reduction in engineering labor as compared to the similar system on the V-22, and at the same time provided a digital artifact to the manufacturing team, which similarly reduced factory development time and labor.”
That doesn’t mean they are done. Model-based engineering is constantly improving, he noted, and the digital thread needs to be continuous and robust throughout the entire life cycle.
“It will not only manage and communicate vehicle performance but also translate back to manufacturing, maintenance and sustainment metrics through the entire supply chain. Digital Enterprise represents continuous feedback from the air vehicle through physics-based analytics, design, virtual validation, manufacturing, readiness, health monitoring, sustainment and fleet awareness.”
Oldroyd explained that “next-generation aviation” will enjoy continued simultaneous maturation of both the product and the process.
“We will utilize an open-architecture digital twin that exchanges real-time data,” he said. “This capability will provide all of the stakeholders—program managers, internal team members, our partners and customers—with near-real-time access to the same information, including engineering analyses, performance characteristics and other relevant metrics.”
One advantage of a highly interactive, enterprise-level digital twin is that the manufacturing space can evolve simultaneously with the design and analysis space—in essence, realizing the digital thread.
To accomplish this, Bell has created a dedicated Manufacturing Technology Center, which Oldroyd said is “a manufacturing innovation environment built upon an Internet of Things-centric digital philosophy.”
To this end, the company is exploring ways to capture sensor-based manufacturing data from CNC equipment and work cells to inform and refine the digital twin.
This will provide feedback about the manufacturing process itself, helping the company optimize productivity, avoid potential quality problems and create a manufacturing history of each aircraft component for the connected enterprise, including the operational teams.
“We will do that.” Oldroyd said. “The digital twin has to be a living organism—one that will adjust to changing circumstances. That way, the manufacturing process will become more and more robust with each passing day.
“Ultimately, work cells can be omniscient elements: They will self-assess. They will inform us when they are not healthy. And they will eventually even take action to become healthy at an environmental level. We’re not finished evolving, but that’s where the ‘digital universe’ is headed.”