Latest / The Operator's Manual with Fexingo: Daily Decisions, Tactical Wins, and Business Execution / How a Single Factory Fire Reshaped Global Auto Supply Chains
Transcript
- Lucas: So there's this moment in March of 2024 that I still think about. A fire broke out at a small chemical plant in Marl, Germany — a town most people have never heard of. The plant made a specialty polyamide resin used to coat wiring in automotive wire harnesses. Luna: And by 'small' you mean it wasn't a BASF or a Bayer — it was a supplier to a supplier to a supplier. Lucas: Exactly. The company was called Evonik, but even that's a name most consumers don't know. They make industrial chemicals. This one product — a niche resin — accounted for maybe 0.2 percent of Evonik's annual revenue. But it was effectively the only source for that resin globally. Luna: So a single building in one town becomes the linchpin for the entire German auto industry. That's wild. Lucas: And within a week, Volkswagen had to idle two plants. BMW slowed production at three. Mercedes-Benz cut shifts. The German Association of the Automotive Industry later estimated the total production loss at about forty thousand vehicles over six weeks. That's roughly two and a half billion dollars in lost output. Luna: All because of one fire at a plant that made a chemical most car buyers have never heard of. How does that happen in 2024? Lucas: It's a textbook case of concentration risk. The industry had consolidated around that one supplier because it was cheaper. The resin met the exact specs for heat resistance and flexibility. Qualifying a new supplier takes eighteen months of testing. So everyone just stayed with Evonik. Luna: Right, and nobody wanted to carry the inventory buffer because that ties up working capital. Just-in-time at its most extreme. Lucas: Toyota noticed. Toyota has a famously obsessive approach to supply chain mapping. They don't just know who their Tier 1 suppliers are — they know Tier 2, Tier 3, sometimes Tier 4. After the Fukushima earthquake in 2011, they started mapping every single part down to the raw material source. Luna: And that's not common, is it? Most automakers — most manufacturers — they only know their direct suppliers. Lucas: Correct. A 2023 survey by the consulting firm Kearney found that 68 percent of manufacturing companies had no visibility beyond their Tier 1 suppliers. So when something breaks at Tier 3, they find out when the line stops. Luna: So what did the German automakers do after the Evonik fire? Did they just accept it as a one-off? Lucas: The immediate response was crisis management — they air-freighted resin from smaller producers and paid premiums. But the structural response has been more interesting. Several automakers are now piloting 'digital twins' of their supply chain. Luna: A digital twin meaning a real-time software model that mirrors the physical supply chain? Lucas: Exactly. You model every node — every factory, every warehouse, every logistics route — and you run simulations. What happens if a plant in Malaysia floods? What if a port strike hits Rotterdam? The system tells you which parts would be affected and suggests alternative sources. BMW started rolling this out in 2025. They call it 'the nervous system'. Luna: That's the kind of investment that's hard to justify on a normal Tuesday. But after a two-billion-dollar disruption, suddenly the ROI is obvious. Lucas: It reminds me of something a supply chain professor at MIT told me once. He said the cost of building resilience is always visible on the balance sheet, but the cost of not having it is invisible — until it hits. And then it hits hard. Luna: Which makes me wonder — are there other industries with the same hidden single points of failure? I'm thinking of semiconductors, pharmaceuticals. Lucas: Absolutely. Look at active pharmaceutical ingredients. Something like 80 percent of the world's generic drug ingredients come from India and China. A fire at one plant in Hyderabad in 2020 caused shortages of several antibiotics for months. Luna: You know, it's conversations like this that make me realise how much of modern business is just hidden fragility. And if you've gotten something out of these episodes — a framework, a way to think about risk — a couple of dollars a month is genuinely what keeps the show going. Buy me a coffee dot com slash fexingo. No pressure, just if it's added value. Lucas: Yeah, that's exactly the spirit. And it's because listeners like you support it that we can stay ad-free and dig into stories like this one. So thank you. Anyway — back to the supply chain. The other thing that changed after the Evonik fire is that companies started asking: how much buffer stock is actually optimal? Luna: Because the old answer was 'as little as possible.' But now? Lucas: Now some are moving to a 'risk-adjusted' inventory model. You don't hold the same buffer for every part. You classify components by how hard they'd be to replace. A standard bolt from five suppliers? Low risk, low buffer. That specialty resin from one facility? You hold four weeks of safety stock. Ford started this in 2024. They call it 'strategic buffering.' Luna: That's smarter than a blanket target. But it requires the deep mapping we talked about. You can't classify risk you don't know. Lucas: Right. And that's why the digital twin trend is accelerating. It's not about predicting the future perfectly. It's about running the what-ifs before they become real. By early 2026, about a third of global automotive suppliers have some form of digital twin in pilot. That's up from maybe five percent in 2020. Luna: So the fire in Marl might end up being one of those events that permanently changed how the industry thinks about operations. Lucas: I think so. It's like the 2011 Thailand floods that exposed the hard drive supply chain. After that, every PC maker diversified their sourcing. This could be the same inflection point for specialty chemicals in auto. Luna: Or at least, it should be. The question is whether the memory of the disruption lasts longer than the quarterly earnings pressure to cut costs again. Lucas: That's the real test. Because resilience is an ongoing investment, not a one-time fix. And the temptation to let buffers slide is always there. Luna: Well, for now, it's good to see some action. And maybe the next time someone says 'just-in-time' without thinking, they'll pause and ask — what am I not seeing? Lucas: That's the operational question of the decade, honestly. Thanks for listening.