SC QUALITY GUILDRolled throughput yield
SCQUALITYGUILD.COM/CALCULATORS/ROLLED-THROUGHPUT-YIELD.HTML
Roll first pass yield across every step to get true rolled throughput yield, normalized yield and the size of the hidden factory.
RTY · NORMALIZED YIELD · DPU · HIDDEN FACTORY · FREE · NOTHING YOU TYPE LEAVES YOUR BROWSER
Q-BotClick for an explanation of this calculator.Yield, DPMO and sigma
RTY · NORMALIZED YIELD · DPU · HIDDEN FACTORYFor educational use. This page is a study aid. The method behind it has far more depth and detail than one page and one tool can cover, and what you see here is not everything the method offers or everything this site offers. For full understanding and application, use the governing standards and texts, and seek further education, training or qualified advice. Results depend on the data you enter, so check them before relying on them.
Final yield is not yield
Most plants report the number of good units leaving the line divided by the number started. It is easy to collect and it is close to useless, because it counts a unit that was reworked three times as a pass.
Rolled throughput yield asks a harder question: what fraction of units get all the way through without being touched? Multiply the first pass yield of every step together and you have it.
The arithmetic is brutal, and that is the point
| Steps | First pass yield at each | Rolled throughput yield | Units touched (1 − RTY) |
|---|---|---|---|
| 5 | 99% | 95.10% | 4.90% |
| 10 | 99% | 90.44% | 9.56% |
| 20 | 99% | 81.79% | 18.21% |
| 10 | 95% | 59.87% | 40.13% |
| 25 | 99.9% | 97.53% | 2.47% |
Ten steps that each run at ninety-nine percent — a number every one of those step owners would call good — deliver 90.4 percent between them. Nearly one unit in ten is touched somewhere. Take the same ten steps down to ninety-five percent each, still not obviously alarming on any single line, and four units in ten need rework.
Nobody owns that number. Each step owner is reporting a healthy figure and the line is bleeding. That gap is the whole reason RTY exists.
The hidden factory
The difference between final yield and rolled throughput yield is work that is really happening and is nowhere in the plan: the touch-up bench, the re-test, the sort, the "just run it again". It consumes labor, floor space, cycle time and material, and it is generally invisible to the cost system because it was never a planned operation.
The useful thing about computing RTY is that it puts a number on that work for the first time. Ten steps at 99 percent means the hidden factory is processing roughly one unit in ten, every day, forever, and nobody has ever costed it.
Normalized yield is the honest average. It is the geometric mean — the yield each step would have to run at to produce the same RTY. Ten steps at a normalized yield of 99 percent give 90.4 percent overall, so quoting "our average yield is 99 percent" and "we lose nearly ten percent" are the same statement. One of them gets acted on.
Where RTY goes wrong in practice
- Rework is counted as a pass. The most common error and the one that makes the whole measure pointless. If a unit was touched, that step did not yield it first pass.
- Steps are defined too coarsely. Calling the whole assembly cell one "step" hides exactly the detail RTY exists to expose. The step boundary should be wherever a unit could be found bad.
- Inspection is treated as a step. Inspection does not make units good; it sorts them. A high yield at an inspection step usually means the inspection is not detecting much.
- Scrapped units silently leave the denominator. If a unit is scrapped at step three and the later steps are reported against what survived, the later yields are flattered.