Latest / The Operator's Manual with Fexingo: Daily Decisions, Tactical Wins, and Business Execution / How One Factory Cut Changeover Time 70 Percent With One SMED Tweak
Transcript
- Lucas: I want to talk about a manufacturing concept that sounds like a magic trick but is actually just really good process design: Single-Minute Exchange of Die, or SMED, which Toyota popularized in the 1970s. Luna: SMED — the idea that you should be able to change a die in under ten minutes, right? Single-digit minutes. Lucas: Exactly. And the classic example is Toyota cutting a stamping die changeover from four hours down to three minutes. But today I want to talk about a more recent application — a mid-size automotive parts supplier in Ohio that cut their changeover time from 90 minutes to 27 minutes with essentially one tweak. Luna: One tweak? That's a 70 percent reduction. What was the tweak? Lucas: They pre-heated the dies offline. That's it. The core SMED insight is that most changeover time is wasted because operators are doing things that could be done while the machine is still running. In this case, the die was cold when it came off the shelf, so the first several minutes after installation were spent waiting for it to reach operating temperature. They bought a simple industrial oven — cost about twelve thousand dollars — and started pre-heating every die to 350 degrees before the changeover even started. Luna: So they converted an internal setup step — something that could only be done when the machine was stopped — into an external step, something done in parallel. That's textbook SMED. Lucas: Textbook, but rarely executed. And the numbers are worth walking through. Before the change, a typical changeover sequence went like this: machine stops, die cools, operator removes old die, fetches new die from storage, installs it, then waits for it to heat up — that wait was 23 minutes of the 90-minute total. After they added the pre-heat oven, that 23 minutes essentially vanished. The new sequence: machine stops, operator removes old die, installs pre-heated die, machine restarts. The wait was gone. Luna: Did they have to change anything else? Like, did the die handling equipment need upgrading? Lucas: Not really. They used the same crane, same clamps, same alignment pins. The only physical change was the oven and a dedicated cart to move the hot die. The real change was in the scheduling. They had to coordinate which die was needed next and get it in the oven about 40 minutes before the changeover. That meant the production planner had to think one step ahead. Luna: That sounds like a small shift, but in a plant running three shifts, that coordination can break down fast. How did they handle it? Lucas: They created a simple visual board — a whiteboard with magnets — showing the current die, the next die, and the status of the pre-heat oven. The operator on the outgoing shift would confirm the next die and start the pre-heat before handing off. It took about two weeks for the habit to stick, but after that, they had a 98 percent on-time pre-heat rate. Luna: And what was the business impact beyond faster changeovers? Lucas: The plant ran 8 presses, each doing 3 to 4 changeovers per day. So they saved about 23 minutes per changeover, times roughly 28 changeovers per day across all presses. That's about 10.7 hours of additional run time per day. Over a year, that translated to about 2,800 more hours of production. The plant manager estimated that at their average margin, that extra capacity was worth about 1.8 million dollars in additional profit — without hiring anyone or buying a new press. Luna: So it's essentially free capacity. That's the holy grail in manufacturing. Lucas: It really is. And it's why I think this case is worth discussing — not because pre-heating dies is novel, but because the SMED framework is still underused. Most factories I visit have changeover times that are two to three times what they could be, and the fix is almost never expensive equipment. It's usually about reclassifying steps: what are we doing while the machine is stopped that we could do while it's running? Luna: Let me push back a little. Isn't this easier for a plant that makes relatively standardized parts? If you're doing high-mix, low-volume runs, the dies vary a lot and pre-heating every one might not be practical. Lucas: Fair point. But in this case, the plant did have 14 different die families. The pre-heat oven had to accommodate different sizes. They ended up buying two ovens — one for small dies, one for large — and that added about 20 grand to the investment. Still, total capital outlay was under 35 thousand dollars, and they got payback in about three weeks. For high-mix shops, the principle still applies: you just need to standardize the pre-heat temperature and cycle time per die family. Luna: Three-week payback. That's almost unbelievable. Why don't more plants do this? Lucas: Two reasons. First, most plants don't measure changeover time precisely. They think it takes 90 minutes because that's what the shift log says, but they haven't broken it down into internal versus external steps. Second, there's a cultural bias that changeover is just downtime you have to accept. Operators and managers alike treat it as a fixed cost. SMED challenges that assumption. Luna: So the first step is just watching the changeover with a stopwatch and categorizing every action. Lucas: Exactly. In this Ohio plant, the operations manager spent a day videotaping changeovers. She played the footage back with the team and asked: 'Why are we waiting for the die to heat? Why are we walking to the storage rack? Why are we searching for the right bolts?' Each question led to a small fix. The pre-heat oven was the biggest, but they also moved the die rack closer to the press — saved 4 minutes per changeover — and color-coded the bolts so the operator didn't have to guess which size to use. All told, they trimmed 70 percent. Luna: That's a great example of the 'go to the gemba' principle — going to the actual place where work happens. The manager wasn't in her office looking at spreadsheets. Lucas: If today's episode was actually useful to you — maybe you're in operations or manufacturing, or you're just someone who thinks about process — these conversations stay ad-free because of listener support. If you want to help keep that going, you can buy me a coffee at buy me a coffee dot com slash fexingo. No pressure, just a way to say this was worth your time. Luna: Yeah, I second that. It's a small gesture that lets us keep digging into cases like this one. Lucas: So back to that Ohio plant — once they had the pre-heat process down, they started looking at other changeover steps. The next target was the alignment procedure, which took about 12 minutes. They found that by installing guide rails on the press bed, they could reduce alignment time to under 3 minutes. Luna: So they kept iterating. That's the continuous improvement loop. Lucas: Right. And within six months, they got changeover down to 18 minutes. Not quite single-minute, but from 90 to 18 is an 80 percent reduction. And the plant manager told me the next target is 10 minutes. They think they can get there by adding quick-release clamps and a die cart with powered rollers. Luna: What about the human side? Did operators resist the changes? Lucas: At first, yes. Some operators felt that faster changeovers meant more pressure — they'd be expected to do more work in the same shift. But the plant manager framed it differently: faster changeovers mean less idle time waiting for the die to heat, and more time actually running parts, which is how operators earn production bonuses. Once they saw the bonus checks go up, the resistance faded. Luna: So the incentives were aligned. That's crucial. Lucas: Absolutely. And it's a reminder that operational improvements don't happen in a vacuum. You have to address the human factors. The plant also started a 'changeover kaizen' team — operators from all three shifts met monthly to review the latest changeover times and suggest improvements. Within a year, the team had contributed 14 ideas, 11 of which were implemented. Luna: That's the kind of ownership that sustains improvement long after the consultant leaves. Lucas: Exactly. So the takeaway for me is that SMED is still one of the highest-leverage tools in manufacturing, but it requires a willingness to question every minute of downtime. The Ohio plant didn't invent anything new. They just applied a 50-year-old idea with discipline and a cheap oven. Luna: And they got 1.8 million dollars in return. That's a pretty good ROI. Lucas: Yeah. So I'll leave listeners with this: what's your equivalent of the pre-heated die? What step in your process are you waiting for that could be done in advance? Luna: That's a great question. I know I'm going to look at my own workflows differently after this.