Estimated reading time: 3 minutes

Nolan’s Notes: Digital Twin in Manufacturing

This month, we look at digital twins in manufacturing. Normally, when we say “digital twin,” we think of design data. But that’s not what we’re talking about; this time, we’re discussing a digital twin for your factory itself. Our discussion follows along with the idea that business operations themselves can have a digital twin. Now why does this even matter?

The folks over at themagical.nl have developed some interesting SIM games based on Disney rides. The idea is that you are the ride operator, and your job is to keep the ride moving smoothly and efficiently even as park guests queue up unpredictably. It’s not as easy as you might think.

Themagical.nl currently shows 14 different ride operator simulations on their website. Naturally, each ride has its own picadilloes to account for; knowing how to operate one ride does not necessarily mean you’re qualified for the others.

While cute and entertaining, I see quite a lot of similarities here. These ride simulators are, without any doubt, digital twins for real rides in real Disney parks. It’s easy, from the perspective of the operator, to view a ride as an assembly line. Guests run through a conveyor belt of experiences much like printed circuit boards run through a conveyor belt of assembly operations. In both cases, operators are charged with keeping those conveyor belts running efficiently and continually, while faced with constantly changing conditions in the input queue.

As a game player, if I could, say, set up certain conditions in the game, and then launch the simulation to test possible solutions, I’ could try strategies to optimize the ride’s operations. If I could reconstruct the ride, I would test optimizations in flow as well as guest experience. I hope you understand my line of thinking: A similar digital twin of your manufacturing floor would allow your team to explore ways to optimize operations. That might be in the form of determining which of your lines is the best choice for a particular job, or it might mean adjusting throughput rates to account for one particularly time-consuming step. It could even mean exploring changes to the setup of your line to decrease operator efforts. There are so many ways to use this kind of digital twin.

It's a freeing feeling, I’m sure, to experiment on the simulator, especially when it feels so risky to try those experiments directly on the manufacturing floor.

I’m reminded of a story from the NASA Apollo missions. Armstrong and Aldrin were working long hours in the lunar lander simulator. Now, in those days, the simulator was almost entirely a mechanical contraption, not necessarily a digital thing, but bear with me. Occasionally, as a landing attempt started to spin out of control, Aldrin would ask pilot Armstrong to abort the attempt. Armstrong, it’s been said, often wouldn’t comply, and would try to recover. Those situations frequently turned into crashes. Apparently, Aldrin took issue with those failed attempts and talked to Armstrong, whose reply was something like, “When we’re up there, we’ll have one chance. We need to know exactly where the line is between success and failure. We can only learn that by crashing in the simulator.” This interaction was dramatized in Tom Hanks’ HBO docudrama series from the late 1990s, “From the Earth to the Moon.” If you’ve not watched that series, it’s well worth your time.

And that’s my point: It’s in the simulator where you can safely explore the boundary conditions and understand your processes better.

In this issue of SMT007 Magazine, we introduce our topic through an interview with Don Kinard, a senior fellow at Lockheed Martin, and one of their experts on digital twin in the manufacturing space. If you haven’t crossed paths with Don, he’s been presenting occasionally at conferences on the different levels of digital twin in industry. Columnist Michael Ford weighs in on manufacturing simulation and, true to form for Michael, he’s looking beyond the horizon with respect to factory simulation; it’s a good read. We also connect with Phil Voglewede at Marquette University’s Omron Advanced Automation Lab, as well as Cogiscan’s Sylvain Perron. And we bring you perspective from digital twin players such as Critical Manufacturing, Instrumental, and Cybord.

All the issues of SMT007 Magazine are a joy to shepherd, and occasionally one particular issue will somehow stand above the rest. Maybe the content was particularly educational or compelling. For me, this is one of those standout issues. I thoroughly enjoyed chasing down these stories and I sincerely hope you enjoy reading them just as much.

This column originally appears in the August 2023 issue of SMT007 Magazine.