Why is first-pass yield the headline rather than the number we report?

Because it is the number that moves when the process improves, and the number that predicts cost. Final yield on most established lines is stable and high, not because the process is good but because rework absorbs whatever the process does. Two plants reporting 96% can have first-pass yields of 95% and 78%, and their costs will be nothing alike. Final yield is still here, still labelled, and still what the gap-to-target row measures, because that is how targets are written.

What is a hidden factory?

The unplanned production capacity consumed by fixing things. It has people, floor space, material and schedule, and it appears in no routing, no standard cost and no capacity plan, so it looks free right up until it becomes the constraint. A rework loop that recovers 10% of panels is a hidden factory running at a tenth of the visible one's volume, and the yield number the earlier version reported was specifically the number that concealed it.

How do I count a panel reworked twice?

As one recovered panel. Yield counts panels, and a panel exists once no matter how many times it goes round. The rework effort is genuinely double, and that belongs on the rework cost page where operations and hours are the unit. Counting it twice here would push recovered plus accepted above panels started, which this page treats as a blocking error precisely because it means the two units have been mixed.

What about panels shipped on a concession?

Decide once and stay consistent. A concession is not a first-pass acceptance, and calling it one defeats the purpose of the page. Most operations count it as recovered, on the grounds that it shipped and consumed engineering time to get there. Counting it as scrapped is also defensible if concessions are rare and treated as failures. What you cannot do is let it drift between the two, because the trend then measures paperwork rather than process.

Our first-pass yield is low but nothing gets scrapped. Is that a problem?

Yes, and it is the most commonly tolerated one. Zero scrap with low first-pass yield means the rework loop is catching everything, which is a real capability and also a real cost. The material is saved; the labor is not. It also caps the line: rework capacity is finite, and when volume rises the loop saturates and scrap appears suddenly rather than gradually. Pricing the loop on the rework cost page usually makes the case for fixing the front end.

Does this measure acoustic performance?

No. It measures how often panels are accepted, whatever accepted means in your inspection plan. A panel can be dimensionally perfect and still miss its absorption or transmission rating, which is a design and material question rather than a fabrication one. If acoustic testing is part of acceptance, its failures belong in these counts; if it is a separate qualification step, it is not covered here.

Acoustic, Noise, Vibration & NVH Products calculator

Acoustic Panel First-Pass and Final Yield Calculator

Measure acoustic panel yield the way it actually behaves, by separating panels that came out right the first time from panels rework rescued. Enter the panels started, how many passed with no rework, how many were recovered after rework, and the target the plan is written against. The calculator returns first-pass yield, final yield, the rework load, and the gap to target. Both yield numbers matter and they answer different questions: the customer experiences the final one, and the cost is driven by the first.

What this calculator does

Separate the panels that came out right the first time from the ones rework recovered, so the yield number shows the hidden factory instead of hiding it.
Use it for finding out whether a comfortable yield number is hiding a rework loop, setting a first-pass target where only a final-yield target exists today, quantifying the rework load before pricing it on the rework cost page, comparing two panel designs on how often they come out right rather than on final acceptance, reviewing whether a process change moved first-pass yield or only shifted work into rework.
Separate the panels that came out right the first time from the ones rework recovered, so the yield number shows the hidden factory instead of hiding it.

Formula used

First-pass yield = accepted first time ÷ panels started × 100
Final yield = (accepted first time + recovered by rework) ÷ panels started × 100
Rework load = recovered by rework ÷ panels started × 100
Panels scrapped = panels started − accepted first time − recovered by rework
Points short of target = target − final yield

Inputs explained

Panels started: Every panel the line began over the period, including ones later scrapped.
Accepted first time: Panels that passed inspection with no rework at all. If the traveller does not record this, the count of rework tickets subtracted from acceptances is a good proxy.
Recovered by rework: Panels that failed first inspection, were reworked, and eventually shipped. Count the panels, not the rework operations: a panel round twice is still one panel.
Target yield: The yield the plan is written against. Measured against final yield, matching how the target is normally quoted.

How to use the result

Best suited to finding out whether a comfortable yield number is hiding a rework loop, setting a first-pass target where only a final-yield target exists today, quantifying the rework load before pricing it on the rework cost page, comparing two panel designs on how often they come out right rather than on final acceptance, reviewing whether a process change moved first-pass yield or only shifted work into rework.
Counts panels, not cost. A panel reworked for two hours and one reworked for five minutes are identical here. Says nothing about defect causes or where in the process they arise. Does not model concessions. A panel shipped under a deviation is neither cleanly accepted nor scrapped and has to be assigned to one. Averaging a mixed panel run hides the difficult panel inside the easy one. Not an acoustic performance measure. A panel can pass dimensional inspection and still miss its absorption rating.

Common questions

Why is first-pass yield the headline rather than the number we report? Because it is the number that moves when the process improves, and the number that predicts cost. Final yield on most established lines is stable and high, not because the process is good but because rework absorbs whatever the process does. Two plants reporting 96% can have first-pass yields of 95% and 78%, and their costs will be nothing alike. Final yield is still here, still labelled, and still what the gap-to-target row measures, because that is how targets are written.
What is a hidden factory? The unplanned production capacity consumed by fixing things. It has people, floor space, material and schedule, and it appears in no routing, no standard cost and no capacity plan, so it looks free right up until it becomes the constraint. A rework loop that recovers 10% of panels is a hidden factory running at a tenth of the visible one's volume, and the yield number the earlier version reported was specifically the number that concealed it.
How do I count a panel reworked twice? As one recovered panel. Yield counts panels, and a panel exists once no matter how many times it goes round. The rework effort is genuinely double, and that belongs on the rework cost page where operations and hours are the unit. Counting it twice here would push recovered plus accepted above panels started, which this page treats as a blocking error precisely because it means the two units have been mixed.
What about panels shipped on a concession? Decide once and stay consistent. A concession is not a first-pass acceptance, and calling it one defeats the purpose of the page. Most operations count it as recovered, on the grounds that it shipped and consumed engineering time to get there. Counting it as scrapped is also defensible if concessions are rare and treated as failures. What you cannot do is let it drift between the two, because the trend then measures paperwork rather than process.
Our first-pass yield is low but nothing gets scrapped. Is that a problem? Yes, and it is the most commonly tolerated one. Zero scrap with low first-pass yield means the rework loop is catching everything, which is a real capability and also a real cost. The material is saved; the labor is not. It also caps the line: rework capacity is finite, and when volume rises the loop saturates and scrap appears suddenly rather than gradually. Pricing the loop on the rework cost page usually makes the case for fixing the front end.

Last reviewed 2026-08-25.