Setup time
The hours a work center spends changing over to run a job, earning nothing, and a cost that belongs to the release quantity that triggered it.
The argument underneath setup time is always about the denominator. A 6 hour changeover against a 50,000 piece release adds fractions of a cent to each piece and vanishes into the line price. The same 6 hours against a 2,000 piece trial order is a visible per-piece cost that can double the quote. So when a buyer asks why the price at 2 volumes differs so sharply, the honest answer is usually setup arithmetic rather than appetite, and the estimator who can show the division wins the conversation.
Where it costs money is when it lives nowhere. Setup hours have to be paid for either inside the work center’s utilization, as hours the machine honestly does not produce, or on the quote as an amortised line. The inherited failure on many floors is that they appear in neither place, counted as productive capacity in the rate and priced on no line of the estimate. Batch sizes have been shrinking for years across the industry, which means changeovers happen more often against the same rate tables, and the leak widens a little every season.
Setup also hides skill. The first pieces after a changeover on tight work were always going to be scrap, and the time spent proving the first good piece belongs to the setup even though it looks like running. A floor that tracks setup honestly, wheel dress to first good piece, quotes from a truer number than one that starts the clock when the machine first cycles.
The release schedule discipline, and why the same part at 2 quantities is 2 different quotes, is worked through with real figures in how to quote a machined part.