Digital Twins, Threads, and Tapestries - oh my

A couple of weeks ago, I attended the CIMdata PLM Road Map & Eurostep PDT North America 2023 event in Washington DC and, I’ve got to say, it was terrific. A great speaker lineup, high-level content, vendor neutrality, and no sales pitches — a refreshing event in a world too filled with commercials.

The event’s theme was The Digital Thread in a Heterogeneous, Extended Enterprise Reality. Of course, everyone had a slightly different definition for “digital thread,” and speakers took it to mean everything from CAD+CAE+PLM+++ to how to implement a global maintenance strategy at the US Navy. No one defined it as “I have a CAD model; I do digital twins,” which I took to mean that these enterprises had a bigger vision and value the digital asset as much as the physical.

Not so fast, said one attendee. He feels that his employer, an aerospace and defense mega-corp, sees digital deliverables only as the means to creating the physical object, with no value of their own. In that perspective, creating these digital models, drawings, etc., is a necessary evil but not worth investing in. And that leads to a self-fulfilling loop: the perceived benefits of these digital assets stop when the physical things are delivered to the buyer, which effectively kills exploring their benefits beyond that point. They're creating the digital models, etc., anyway; not using them to their fullest is a waste. More on this below.

My favorite presentation was the nerdiest: among many other things, this project updated a complicated two-PLM scheme to using a single PLM system by revolutionizing how they look at Bills of Material (BOMs). A BOM is a list of all the parts that make up an assembly: 1x partA, 1x partB, 3x partC, etc. When a BOM changes, it’s wholly recreated: 2x partZ, 1xpartB, 3xpartC, and so on. What if we didn’t do it that way but instead just noted the change: remove 1x partA but add in 2x partZ … It’s more straightforward and concise. It can be spun backward and forward to note how the BOM evolved. It’s a simple-seeming change that has enormous ramifications for an enterprise with millions of BOMs. Elegant and genius-clever.

One surprise to me was the pace of change reported by these speakers. Most came from large aerospace and defense contractors who have, over time, built incredibly complex and overlapping technology landscapes, with some legacy systems still bearing significant load and, therefore, hard to change. One speaker has been trying to replace an ancient PLM system for years but repeatedly failed because of customizations that users don’t want to give up. Concern for user engagement meets IT realities -- not easy to resolve.

Many of the companies I deal with aren’t that — they’re smaller and perhaps hungrier and so more agile. They’re looking for the opportunities created by the innovative use of new technologies. They look at these technologies as a way of growing their businesses rather than using them to (grudgingly) meet customer requirements to “be more digital.” But I still maintain that trying a small implementation of a digital thread concept on a project or a small division to see what's possible is a good idea, even for these huge enterprises, and perhaps can lead to incremental changes.

Back to the fellow who said his company doesn’t value digital assets. Many of these speakers said their companies were motivated to “do something digital” by their customers. One started on a digitalization quest when the customer required MBSE (model-based systems engineering) methodologies in a contract award. Another had a collapsing PLM infrastructure and couldn’t meet customer requirements, so ... It’s good that something started these processes sooner rather than later, but working on such significant changes in a reactive mode adds stress to an already challenged IT environment. A more thought-out, internally-motivated catalyst might do better. Moral: start thinking about this now, before you have to.

There was one other inescapable conclusion from the event: waving one’s hands and saying, “The digital thread connects all things to all things,” is ridiculous if there isn’t a way to actually do that. Twins won’t happen at scale until data can be freed from one system to be used in another context with accuracy and completeness (and at a reasonable cost). Standards are not my favorite topic, but they are essential if digital twins/threads/tapestries are to become a reality for more enterprises. Data stuck in silos is useless, and perhaps one reason for the enterprise that doesn't value digital: it's hard to get at, so let's not bother.

And one last thing: one of the speakers was the President of Engineers without Borders, USA . This organization teams up with communities and non-governmental organizations to do the engineering part of public works projects worldwide — bringing clean drinking water to a village, for example. Their work is incredibly worthy, and much of it is done by engineering students. I’ve been fortunate to meet some of these budding engineers and was impressed by their poise, interest, and engagement. They’d be terrific employees! And they need mentors, seasoned engineers who can guide student teams If you have the chance, please consider donating to Engineers Without Borders, hiring their student volunteers, and offering to mentor -- it's a double win for the student and the communities where the projects take place.

Thanks to the team for putting on such a great event and to all the speakers who prepared and delivered terrific presentations.

---

16 May: Updated to reflect that this was actually a joint event hosted by CIMdata and Eurostep. I apologize for the oversight.