Latest / The Manufacturing Podcast with Fexingo: Factories, Supply Chains, and Industrial Business / How a Taiwan Factory Uses Robots to Make Sneakers
Transcript
- Lucas: So I just got back from a factory in central Taiwan that makes sneakers — and I mean, makes them almost entirely with robots. Luna: Wait, like — the whole shoe? From start to finish? Lucas: Pretty much. This plant is run by a company called YF Shoe, which is a contract manufacturer for brands like Nike and Adidas. They've been around for thirty years, but two years ago they decided to build a fully automated line. Luna: And they actually pulled it off? I know a lot of factories have tried, but shoes are notoriously hard to automate because of all the flexible materials and complex shapes. Lucas: Exactly. But YF partnered with a robotics firm in Japan called Fanuc and spent about $50 million on the line. They now produce a pair of sneakers in under 20 minutes — start to finish. Luna: Twenty minutes? That's incredible. Traditionally, a pair of sneakers takes hours of manual work, spread across dozens of workers. Lucas: Right. The old way: you have people cutting fabric, stitching uppers, gluing soles, assembling. Each step is done by hand. YF's automated line uses robotic arms for cutting, computer vision for aligning parts, and automated glue dispensers that apply exactly the right amount. Luna: And the computer vision — that's what handles the variability? Like, the robot can adjust if the fabric stretches? Lucas: Exactly. The vision system scans each piece as it comes in, compares it to the CAD model, and tells the robot arm exactly where to place it. That's the breakthrough — the system can adapt to slight variations in material. That's what made shoe automation so hard before. Luna: So what's the actual output? How many pairs per day? Lucas: The line runs about 500 pairs per shift. That's roughly what a manual line with 60 workers would produce. But the robot line runs with only 10 human operators — mostly for supervision and maintenance. Luna: So labor costs drop by what — 80 percent? Lucas: About 75 to 80 percent, depending on local wages. But that's not the only saving. The precision reduces waste. Manual cutting wastes about 15 percent of the leather or synthetic fabric. The robot line wastes under 5 percent. For a factory using tons of material, that's huge. Luna: And I imagine the consistency is better too. No human error in gluing or stitching. Lucas: Right. Defect rates dropped from around 3 percent to under 0.5 percent. That means fewer returns, less rework, and better brand reputation. Luna: But $50 million is a big upfront cost. How long is the payback period? Lucas: YF told me they expect to break even in about four years. That's based on labor savings and waste reduction. But they also benefit from faster changeovers — they can switch between different shoe models in about 15 minutes, versus hours on a manual line. Luna: That's a game-changer for small batch production. Brands want to release limited editions quickly. Lucas: Exactly. The line can handle up to 20 different models per day. So YF can offer faster turnaround to their clients, which is a competitive advantage. Luna: I'm curious about the human side. What happened to the 50 workers per shift who were replaced? Lucas: YF retrained many of them for other roles — quality control, maintenance, programming. But not everyone. The factory has about 200 fewer workers now than two years ago. The company says they've been able to absorb some through attrition, but it's not a clean story. Luna: And that's the tension, right? The factory is more efficient, but the local community loses jobs. Lucas: Exactly. YF is in a small city called Douliu, where manufacturing is the main employer. The mayor has been pushing the company to invest in retraining programs. YF does offer classes, but the skill gap is real. Luna: Do you think this model will spread to other shoe factories? Or is it only for high-end production? Lucas: For now, it's mostly for mid-to-high-end sneakers. The robots can handle leather and synthetic materials, but they still struggle with very soft textiles used in some athletic shoes. And the upfront cost is prohibitive for smaller factories. Luna: So the big players — Nike, Adidas, Puma — they'll adopt it, but the smaller contract manufacturers might get left behind. Lucas: That's the trend. In fact, Nike has been investing in automated manufacturing for years. They have a process called Flyknit that uses a single-piece upper, which reduces assembly steps. But full automation of the entire shoe is still rare. Luna: YF seems ahead of the curve. What's next for them? Lucas: They're building a second automated line that will handle a wider range of materials, including recycled plastics. They're also experimenting with 3D printing for midsoles. The goal is to produce a shoe that is 95 percent automated by 2028. Luna: That's ambitious. But if they pull it off, it could reshape the entire supply chain. Shoes could be made closer to the consumer, not just in Asia. Lucas: Exactly. Imagine a sneaker factory in the US or Europe that can produce small batches on demand. That would slash shipping costs and lead times. YF's model is a proof of concept. Luna: It's fascinating. And it makes you think about all the other industries that are ripe for this kind of transformation. Lucas: Look, this is exactly the kind of conversation we love having on this show. And it's only possible because we don't run ads. We don't have sponsors telling us what to cover or how to frame it. We just follow the story. Luna: And if these conversations have moved your work forward or given you a new angle on your own industry, that's the whole point. Lucas: Exactly. So if you want to support that independence, the link is buy me a coffee dot com slash fexingo. No pressure, just a way to keep this going ad-free. Luna: And it genuinely helps. We've had listeners tell us they use these episodes in team meetings or to pitch new ideas. That's the kind of impact we want. Lucas: Alright, back to the factory. One thing I didn't mention: YF's automated line also tracks every step with RFID tags on each shoe. So the customer — Nike or whoever — can see exactly where each pair is in production. Luna: That kind of traceability is becoming a requirement for big brands. They want to prove their supply chain is ethical and efficient. Lucas: Right. And because the robots handle the gluing, there's less exposure to toxic chemicals for workers. That's a big selling point for sustainability reports. Luna: So the factory isn't just faster and cheaper — it's also cleaner and more transparent. Lucas: Exactly. And I think that's the real story here. Automation isn't just about replacing people. It's about rethinking the entire manufacturing process to be better in every dimension. Luna: But we can't ignore the cost to workers. Those 200 people lost their jobs. Lucas: No, we can't. And YF isn't a charity. They made a business decision. The question is: what's the social safety net? In Taiwan, the government has been slow to fund retraining. That's a policy gap we'll need to address as automation spreads. Luna: It's a reminder that technology moves faster than policy. But that's a topic for another episode. Lucas: True. For now, I'm just impressed by what YF built. I walked out of that factory thinking, 'This is what the future looks like.' Luna: And it's happening right now, not in some distant future. Lucas: Right. So if you're in manufacturing, pay attention. The robots are coming, and they're getting smarter every year.