When a supplier arrives short on a production morning

A restaurant discovering it is short on salmon removes it from the specials board. Nobody who wanted salmon knows they were disappointed, because they never ordered it.

A meal-plan kitchen short on salmon has forty customers who specifically selected it, each expecting it in a box with their name on it. There is no quiet way out — every option is visible to somebody, and the only real choice is which visible option you take.

That is why supplier failure is a decision problem with a clock on it rather than a procurement problem to resolve later.

The reflex under pressure is to swap in whatever is available and similar. That reflex is exactly how an allergen reaches someone who reacts to it, because the substitution happens on the kitchen floor rather than in the system that holds the dietary records.

The discipline worth building is that any dish-level substitution is re-validated against affected customers before it is cooked. Ingredient-level substitution within a dish needs the same check if it introduces anything new.

This is one of the few operational rules worth being genuinely inflexible about, because the failure mode is not a refund.

Two things reduce the pain, and both are cheap in advance. First, know which ingredients have no acceptable substitute and which suppliers have no realistic backup — that list is usually shorter than expected and tells you where to hold buffer.

Second, decide the credit rule in advance. Deciding what a shorted meal is worth while forty customers are waiting produces inconsistent answers that become disputes later.

A third habit helps: order slightly earlier than the lead time requires for the handful of items that would stop production. The cost is a little working capital; the benefit is discovering a shortfall while there is still time to act.