Latest / The Operator's Manual with Fexingo: Daily Decisions, Tactical Wins, and Business Execution / How One Factory Cut Defects 90 Percent With a Single Board Change
Transcript
- Lucas: In 2019, Toyota's Georgetown, Kentucky plant was producing roughly 550,000 vehicles a year. One of the most persistent quality headaches was a tiny defect in the door panel seal. Luna: The rubber seal that runs along the edge of the door? Lucas: Exactly. When it doesn't seat perfectly, you get wind noise, water leaks, and eventually customer complaints. The defect rate was running about 1.7 percent — which sounds low, but on half a million cars, that's over 8,500 vehicles with a bad seal. Luna: And at an average repair cost of roughly $270 per car for rework and warranty claims, you're looking at over $2.3 million a year. Lucas: Right. So the plant's quality team put together a cross-functional group — engineers from body assembly, paint, and final assembly — to do a proper root-cause analysis. They'd been tweaking the process for months: adjusting the robot that applies the adhesive, changing the curing temperature, even swapping seal suppliers. Nothing got the rate below 1.5 percent. Luna: So they went back to the floor. 'Go and see,' as Toyota says. Lucas: Yes. They spent three days just watching the line — not the quality check station at the end, but the actual installation step. And what they noticed was subtle. The robot arm that positions the door panel before the seal is applied has a metal guide — a fixturing jig — that holds the door in place. That jig had a two-centimeter gap between the guide surface and the door edge. Luna: Two centimeters. That's less than an inch. Lucas: About three-quarters of an inch. And because of that gap, the door could shift by a millimeter or two during the seal application. That tiny movement was enough to create an uneven bond between the seal and the door frame. The quality check at the end of the line — a visual inspection and a water spray test — was catching the obvious failures, but the marginal ones passed through. Luna: So the fix was to close the gap in the jig. Lucas: They added a two-millimeter-thick metal shim to the guide — essentially a thin strip of steel — to reduce the clearance. Total cost: about $12 for the shim, plus 45 minutes of line downtime to install it. Within two weeks, the defect rate dropped from 1.7 percent to 0.8 percent. By the end of the second month, it was at 0.15 percent. Luna: A 90 percent reduction. For twelve dollars. Lucas: And here's the part that makes this a classic operations story. The plant manager had actually seen the gap himself during a walkthrough eighteen months earlier. He'd noted it in a log, but the engineering team had dismissed it because the jig was within the original equipment manufacturer's tolerance specifications. Luna: So the fix was known — it just wasn't acted on. Lucas: Exactly. The tolerance spec said the gap could be up to three centimeters. The jig was at two, so it passed. But the spec was set by the machine builder, not by the process outcome. Nobody had asked the question: 'Does this spec actually produce a good seal?' Luna: That's the kind of thing that sounds obvious in hindsight. But when you're on the line, you tend to trust the number on the print. Lucas: And that's a really useful lesson. The Georgetown team now has a policy: any time a defect persists beyond three months, they go back and question every spec that touches the part — even if the spec is 'in tolerance.' They call it a 'spec challenge.' Luna: I love that. How many other processes have specs that are technically correct but practically wrong? Lucas: Hundreds, probably. And that's the real takeaway here. The shim itself is trivial. The process of questioning your own assumptions — that's the hard part. Luna: If this kind of concrete, real-world operations thinking is useful to you in your work, it's worth mentioning that these episodes stay ad-free because of listener support. People who find value here can help keep that going at buy me a coffee dot com slash fexingo. Lucas: Yeah, and it genuinely makes a difference. Every contribution lets us keep drilling into these specific cases rather than chasing algorithms or ad revenue. So if today's episode moved your thinking about your own processes, that's the way to keep it coming. Luna: Back to the jig. What happened after the fix went live? Lucas: The plant documented the change in a one-page kaizen sheet — before, after, cost, savings. That sheet got shared across Toyota's North American manufacturing network. Within six months, three other plants found similar jig clearance issues on different assembly steps and applied the same logic. Luna: So the $12 fix turned into a network-wide improvement. Lucas: Right. And the total annualized savings across all four plants came to about $6.8 million. All from one shim. Luna: What about the original plant manager who saw the gap and didn't push it? Lucas: He's still at Georgetown. The company didn't punish him — they used it as a teaching case. At Toyota, the idea is to fix the process, not blame the person. So they changed the spec-review protocol: now, any fixture that's been in service more than three years gets a biannual 'go and see' review by a cross-functional team, not just the maintenance crew. Luna: That's a structural fix. It prevents the same failure from happening again. Lucas: Exactly. And it's a good reminder that most operations breakthroughs aren't about brilliant new technology. They're about looking at something you've looked at a hundred times and finally asking the right question. Luna: So if you're running a process, the question is: what spec are you trusting that might be quietly costing you? Lucas: That's the question. And the answer might be hiding in a two-centimeter gap you've walked past every day.