Operations and Supply Chain

Just-In-Time inventory sounds like pure efficiency. It is — right up until it isn't.

The core idea of JIT is straightforward: instead of holding weeks or months of stock in a warehouse, you receive exactly what you need, exactly when you need it. No warehousing costs, no capital tied up in idle inventory, no obsolete stock. Toyota pioneered this in the 1970s and achieved remarkable efficiency gains. The problem is that the system works because of everything surrounding the inventory policy — deep supplier relationships, geographically concentrated supply chains, frequent small deliveries, and near-zero defect rates. Strip those out and what you're left with is not efficient JIT. It's fragile JIT: a system with no buffer that fails catastrophically the moment anything goes wrong upstream.

The expected-value math on JIT often looks favorable: occasional line stops versus continuous warehousing costs. But this comparison misses what statisticians call tail risk. A line stop isn't a normally distributed event. A supplier fire, a port shutdown, a quality recall — these produce shutdowns measured in weeks, not hours. The expected annual cost calculation smooths over this. A $300M revenue loss from a two-week shutdown is not the same category of event as the model's "$800K annual expected line-stop cost" implies.

Now here's where it gets harder in practice — because the CFO's expected-value math looks right on paper, and the COO's nervousness doesn't have numbers attached to it yet.