- First pass yield is units completed correctly without rework, divided by units started.
- Rolled throughput yield multiplies the first pass yields of every step, and is usually far lower than anyone expects.
- Final yield hides rework. The gap between final yield and first pass yield is the size of your hidden factory.
A line reporting 98% quality may be reworking a fifth of its output before that check. Final yield cannot see it. First pass yield is designed to.
How to calculate it
First pass yield = units that passed without any rework, divided by units that entered the step. The critical word is any: a unit that was adjusted, re-run, touched up or re-tested does not count as a pass, even though it eventually shipped.
Rolled throughput yield
Multiply the first pass yield of each step. Five steps at 95% each gives a rolled throughput yield of about 77%, meaning fewer than four in five units go through the whole process untouched. This is usually the number that changes the conversation, because each individual step looked fine.
What the gap tells you
- A large gap between final yield and first pass yield means substantial rework, which consumes capacity nobody has budgeted.
- Rework is also where variation enters: a reworked unit has a different history from a first-pass one.
- Rework hours are frequently invisible in costing, so the true cost of poor quality is understated.
Measuring it without a new system
Record, at each step, units in, units out first time, and units reworked. Three numbers per step per shift. The main obstacle is cultural rather than technical: rework is often done quietly because reporting it feels like reporting failure. That has to be addressed openly before the numbers mean anything.
What to do with a low figure
Find the worst step, then look for the cause rather than the defect. Rework concentrated at one station is normally a method, tooling or incoming-material problem, not an operator problem, and treating it as the latter guarantees the figure stops being reported honestly.