Ep. 30: Supply Chain Resilience Through Better Data – How Accurate BOMs Impact Material Planning

Supply Chain Resilience Through Better Data: How Accurate Bills of Material Impact Manufacturing
The Integrate Intelligently Podcast — with Jeff Bickler and Scott Bickler (CADTALK)

Jeff: Welcome back to The Integrate Intelligently Podcast. I'm your host, Jeff Bickler, joined by my brother, founder and CEO of CADTALK software, Scott Bickler. Today we've got an interesting, very topical episode: supply chain resilience through better data, and how accurate bills of material impact material planning and manufacturing.

Scott: Good to be with you today. This is an interesting topic — one that's come up again recently, mostly because of new federal policies coming out of the U.S. government. So it's a good time to talk about how those policies affect manufacturing.

Jeff: Absolutely. I think the last time supply chain got this much coverage was during COVID, with all the disruptions we had then.

Scott: Right. I think it'll be interesting to talk about how we help with that — but also how complex everything people buy and own really is. How it all comes together to end up in their Amazon box.

What Is a Supply Chain? A Real-World Example
Jeff: That's a good jumping-off point. I've been working with manufacturing companies for the last decade, but you've been in it for two, maybe three decades. We won't go too far down that road — but maybe give us some examples of what a supply chain actually looks like, since it's not a term we hear explained often.

Scott: I was actually thinking about this before the call. The most relevant day-to-day example is probably your car. You buy a car from Ford or Tesla, but they didn't make everything that goes into it — they bought it. So the first link in the chain is the manufacturer, or really the dealer who bought it from the manufacturer. That's already two links.

That car manufacturer is often mostly doing assembly. They're buying the carburetor, the engine — all these pieces come together, and that's the second link. They're buying those from vendors, which is another link in the chain. Then all of those suppliers are buying steel, castings, electronic components — all the way down, chained together with dependencies, to give you a car you can buy at the dealership. Any disruption in any of those parts keeps the car from shipping.

Here in Cincinnati, during COVID, a whole motor speedway in Kentucky — practically a racetrack — filled up with Toyota trucks missing one chip. They couldn't get it shipped from Asia. Without that one chip, the car doesn't run. You plug the chip in, and they drive off with it. Without it, nothing.

Jeff: No go. Can't do anything with it.

Scott: That gives you an idea of the complexity — all these companies around the planet coming together to make one truck.

Why Globalization Made Supply Chains More Fragile
Jeff: That's something I didn't appreciate until I started talking to manufacturing companies and working at CADTALK. You hear about globalization in the news — and about the breakdown of global supply chains during COVID. What that really means is that manufacturing companies, wherever they're based, get their materials — purchased parts, raw materials, whatever they need — from somewhere else in the world. Sometimes six, seven, eight, ten different countries, all converging on one location for final assembly.

Scott: Absolutely. The world has far more complexity than people realize. Think back to the days of a blacksmith forging something out of ore he dug out of the ground — there weren't many links in that supply chain, so there was less to disrupt. As we create more complex, more diverse products, very few companies are what's called "vertically integrated" — owning the entire supply chain.

In a lot of ways, it's actually good that they don't. It creates less risk, more competition for the same work, and it builds an economy — a little margin gets made at every exchange, which feeds the whole system. Not to get into macroeconomics, but that's basically how economies work: chains like that. It lets people focus on one small piece, optimize it, and the combination of everyone optimizing their piece produces the amazing technology we have today.

Jeff: It also reflects the buildup we've had since the end of World War II — building out global supply chains across Asia, Mexico, the U.S., Canada. Manufacturing companies pull from different countries that specialize, whether it's the right raw material, special technology, or cheaper labor. Then all those manufacturers across the globe come together into one product, like a car.

The more complex the product, the more complex the supply chain. Think about semiconductor manufacturing — tooling from the Netherlands, silicon from another country, manufactured in Taiwan or elsewhere in Asia, or even the U.S. All these complexities have to come together to make one product.

Scott: Absolutely. You can debate whether that's good or bad, or how it got that way — we're not trying to get political here, just educating on how it evolved, and where our small part fits in as integration software connecting engineering and manufacturing systems.

From Engineering Design to the Supply Chain: "Art to Part"
Scott: Everything starts with an engineering design — if it's manufactured, it starts there. We call it "art to part": from the art, the brain of the person who created it, all the way to the physical part.

Jeff: Never heard that term before.

Scott: [laughs] I thought I'd used it before, but okay — that's what we call it. You're taking what's in the engineer's mind and bringing it into reality. Now, engineers are usually designing in the context of things that already exist — standard bearings, standard steel — but sometimes they conceive of entirely new materials or components. Either way, everything comes from some kind of engineering design.

Our job is to streamline the transfer of that engineering design into the supply chain systems — usually an ERP or some downstream system. There's a hard-cost benefit to that speed, which is what most people see. But let's talk about the soft cost: the faster and more accurately we can move that data, the more resilient you actually are to supply chain issues, because you're able to react and change the design to deal with whatever's happening.

How Supply Chain Disruptions Ripple Down to the Engineer
Jeff: Let's step through the scenario. You're an engineer, you design a product, you release it, and it ends up in the ERP. People generate purchase orders to buy the parts. Now — how do supply chain interruptions, tariffs, or any disruption ripple down into the individual's day-to-day work?

Scott: First, if the data isn't transferred accurately — brought over manually, or with mistakes — you're going to buy the wrong things, or the wrong amounts. That's table stakes, and it's the first thing we'd help solve.

The other piece: say you're the engineer reacting to a shortage. Take the car example. You can't get the chips you need — you can't ship cars, can't make profit. Can you source a different chip domestically, or from another country, and retrofit it with an added component? The engineer investigates, figures out that a different chip works with a small modification, and then needs to convey that change into the system — fast. That's how you react to a short-term issue: going from art to part to supply chain very quickly, instead of having every step in the chain become a bottleneck for the information itself.

That's really the core of it: the engineer decides "I can't use X, I need to use Y," and figures out what needs to change to make Y work. That's how you build resilience.

Design is always a game of trade-offs — true in code, true in engineering. I can build something out of the cheapest material, but it might not be as strong. I can build it from an expensive material, but it might not be as common. What am I optimizing for? If the supply chain becomes the major constraint — the bottleneck — I can design differently around that constraint than I could before.

Jeff: So lead times matter — lead times for alternate parts, and how quickly that data gets back to the engineer to make a decision. If we can't manufacture the original design anymore, engineering has to redesign around Part B instead of Part A — sourced from a different, cheaper, or more available supplier. The faster engineering knows that and can adapt, the faster they can push those changes back into the supply chain.

Scott: Right — and think about what happens the moment a supply chain is disrupted: everyone rushes to the same new source. If you had five vendors and now one country's supply is cut off, three of those five vendors, everyone's now competing for the remaining two. The speed at which you can get your parts retrofitted and reordered determines whether you actually get them, because those two vendors are about to see demand double or triple.

Applying Software's "Continuous Integration" Lesson to Manufacturing
Scott: It's interesting — software borrows a lot from manufacturing processes, but here's a case where it goes the other way. In software, there's a practice called continuous integration and continuous deployment. If I find a bug, I want to fix it, verify it, and have it automatically deploy to production through a series of automated gates — because at every manual step in a complex system, you risk reintroducing a problem.

We do this ourselves at CADTALK — automated builds, automated tests. Once we verify a change is working, it can move through deployment very quickly, so our reaction time to a bug or a needed feature is fast.

Apply that same principle to manufacturing: there's a "bug" — a supply chain issue. It goes back to the engineer, we decide on an alternate plan, implement the design change, and then need to move through the deployment steps as fast as possible. That's where CADTALK comes in.

Jeff: Whereas if you don't have a modern ERP, or you're using spreadsheets with no integration and everything's manual — not only is it slower, but the disruption ripples through the whole organization. Every gate becomes even slower than before, because now you're checking spreadsheets, revalidating data, manually grinding through each step with no automation to shorten the timeline.

Scott: Right — imagine a manufacturing company without automated gates. Someone has to make every decision manually, and they're in a rush. People under stress make mistakes. That's the beauty of software: it does the same thing every time, doesn't get in a rush, just runs fast. In software development, this kind of automation is now table stakes — nobody does manual builds and deployments anymore, because the complexity is too high and it's too easy to mess up.

Jeff: Right, and it's not just that the outside world is affecting you — you then have to go through every other internal process, all manual, with no integration, to make those changes. That slows down getting your product to the customer. You may not even get the parts, like Toyota — all that capital tied up in inventory you can't sell.

Scott: That's the "golden sprocket" problem in manufacturing — you have everything you need except one thing, and your money is tied up in inventory that can't be used.

Building Resilience With Alternate Parts, Routes, and Vendors
Jeff: Scott, what are some other examples of how systems working together — or bill of material accuracy — can make engineering and manufacturing companies more resilient against these disruptions?

Scott: Beyond bill accuracy itself — right quantities, right structure — a lot of ERPs support alternate parts, alternate materials, and alternate routings. If you maintain those, you're building in resilience. People often route parts based on the machinery they currently have, which makes sense. But if a machine goes down, or work needs to be outsourced, that's a supply chain disruption too — it just happens to be internal. If you plan around that, you know what to do: laser's down, switch to a router or a torch.

The problem is, a manufacturing engineer evaluating every design doesn't usually have the bandwidth to build out alternate paths for everything. It's a risk-reward calculation — how often will this actually happen versus the time it takes to prepare for it?

That's where AI changes the equation. Not only do you have the design the engineer created, but the system can start inferring and building out alternate paths ahead of time for likely scenarios — so when the laser goes down, the software can already default to the router without a person having to stop and decide.

Jeff: So if your ERP can handle alternate parts, alternate routings, alternate vendors — and you build that data out — you'll be able to respond more quickly to any supply chain disruption, because you've already thought it through.

Scott: Exactly, but the upfront cost of figuring all that out manually is high. Someone in supply chain, engineering, or manufacturing has to decide what the alternates are and enter them all — and that effort compounds. So realistically, people often just pick one way to do it, because they need the part ordered and made. They don't mitigate the disruption risk, because it doesn't happen often enough to justify the time, or they're just under pressure to get it into the system.

Jeff: So it's really about risk mitigation strategy — but where CADTALK comes in is making it easier for engineers to generate those multiple BOMs, multiple routes, multiple vendor options, much faster.

Scott: Right. What used to take eight hours to enter all the alternates — just to cover your bases — isn't nearly as big a hurdle anymore. Our integration and design tooling can build those out quickly and push them into the system, which lowers the barrier to actually doing the work.

A Real Example: Standardizing Sheet Metal Finishes
Scott: I've got a concrete example from our nesting integration work. Years ago, we made a lot of laser-cut parts in stainless steel with different finishes — a "number three" finish, which is a polished grain on one side, versus a "number four" finish, polished on both sides. If a part isn't visible from one side, you don't need that side polished — so specifying number three instead of number four can save material cost.

But when you're laser cutting, the more small parts you can fit into a nest, the higher your material yield. So we decided: anything under a certain size, if it was specified as number three, could also be made from number four material — because number four covers both cases, but not the reverse. That let us fill out a nest with parts that would otherwise sit idle, slightly increasing per-part material cost but reducing lead time and avoiding the golden-sprocket problem.

That's adjusting to a supply chain issue in real time — in that case, an internal one: which finish happened to be on the laser at that moment.

Jeff: Same logic applies externally. If I can't get one material, but I can get a slightly more expensive alternate, I at least have options. But you have to have already built out those alternates and those scenarios — otherwise you're figuring it out on the fly, and that's slow.

Scott: Right, and that's the world AI is taking us into. Decisions that used to require a person to analyze and decide can increasingly happen almost instantly — "this is basically equivalent, go ahead and make the change." When the barrier to creating alternates drops to near zero, the whole calculation changes: you'd just do it, because it's no longer costly to prepare for.

Resilient vs. Antifragile: The Real Goal
Jeff: So the barrier to doing that work — which people skip for good reasons today — disappears, and the system can just do it automatically. That makes things more resilient. Though at the end of the day, disruptions still happen. The question is whether it's a speed bump or a break in the road.

Scott: Right — and that comes down to how resilient you are, and ideally, how antifragile you are. Resilience is how quickly you return to your steady state after something goes wrong. Antifragile means you come back to a better state than before — you find out the disruption actually pushed you toward something cheaper or better.

Jeff: That reminds me of a story you've told before, about standardizing sheet metal sizes almost by accident, and how it ended up making things more antifragile.

Scott: Yeah — we standardized on two sheet sizes because we knew supply was favorable for that material. But because we were cutting on lasers, standardizing meant better yield across the board. Even though some individual parts cost slightly more, overall cost went down. A series of seemingly suboptimal decisions at the micro level led to a great outcome at the macro level.

Jeff: That's why you brought it up — supply chain disruptions, if you're antifragile, don't just get you back to normal. They can actually produce an improvement. But only if you're ready for them and your system is built that way.

Scott: Exactly — like the market. When it drops, there's opportunity for those positioned to take advantage of it. If you're not in that position, you won't just miss the upside — you'll fall further behind while competitors who are ready take advantage.

Key Takeaways
Jeff: This has been a great conversation. A few things I'm taking away: supply chains are more global — and more complex — than most people realize, and disruptions anywhere in that chain can stop a product cold. I also hadn't thought about supply chains as something that can be internal, not just global sourcing — it's really any process of getting what you need to manufacture, whether that source is across the world or across the shop floor. And if you're antifragile, disruption can leave you stronger than before.

Scott: Thanks for listening. Whatever happens with tariffs or other disruptions, we think it's important to focus on what you can control to become more antifragile and resilient — and that's what we're trying to help our customers do: improve their processes so that when disruptions hit, they're ready to respond.

Head to cadtalk.com to read our blog, learn more about supply chains and nesting, and subscribe to the podcast on YouTube, Spotify, or wherever you listen. Feel free to write in with topics you'd like to hear Jeff and me cover.

Jeff: Anything else before we wrap up?

Scott: No, I think that was great.

Jeff: Thanks, everyone — we'll see you next time.