First-pass yield (FPY) is the percentage of units that meet requirements without a later rework pass. The basic formula is first-pass-good units divided by the units entering the batch, multiplied by 100. Final or blended yield can be higher because it may include units accepted after rework. Rolled throughput yield (RTY) answers a different question: how first-pass performance compounds across several process steps.
If you only report the final number, a batch can look healthy while the process has required another inspection, correction, or production loop. This page shows the three metrics side by side and gives you a small batch log to keep the rework visible.
First-pass yield: definition and formula
First-pass yield measures what passed the first time. It does not count a unit as first-pass good if that unit needed correction or another processing pass before release. SAP’s manufacturing guidance separates first-pass output from later-pass or retest output, and the same boundary is used in this article.
Use this formula:
FPY = first-pass-good units / units entering the batch x 100
For example, if 900 of 1,000 units meet requirements without rework:
FPY = 900 / 1,000 x 100 = 90 percent
The denominator is the first control point. Write down what entered the process before you calculate the percentage. If one source uses units started, another uses expected output, and a third uses final accepted units, those values should not be compared as if they were the same metric.
There is no universal “good” FPY percentage to copy from another plant. Set a plant-specific baseline by comparing like-for-like products, process conditions, defect definitions, customer requirements, and measurement boundaries. A percentage is useful when the team can explain what it includes and what changed.
Final or blended yield and what it hides
Final or blended yield is useful when the question is how many units were ultimately accepted. In this article, define it as final acceptable units divided by the same original batch input. State explicitly whether the final count includes units that were reworked.
That distinction matters because a reworked unit can be acceptable at the end and still be a first-pass failure. The final number tells you where the batch ended. FPY tells you how much of the batch reached that point without another loop. ASQ’s cost-of-quality guidance recognizes scrap and rework as internal failure cost categories, so the final output number is not the whole process story.
The supplied video segment uses an illustrative contrast of about 90 percent first-pass yield and about 98 percent after rework. The 98 percent result is not evidence that the batch performed at 98 percent on the first attempt. It means the process recovered some of the loss through additional work.
For the wider cost boundary, see how to measure the hidden cost of manufacturing waste and rework.
First-time yield versus first-pass yield
First-time yield (FTY) is often used to describe first-attempt performance, and some sources use it interchangeably with FPY. “First time right” is also commonly used for first-attempt acceptance, but the local definition can vary. Terminology varies by plant and software system, so state the definition and denominator instead of assuming the labels are universal synonyms. This article uses FPY for units that meet requirements without a later rework pass.
Do not use “yield” without a boundary. In a manufacturing setting, it may refer to material conversion, expected output, final accepted output, first-pass performance, or a standard-based expectation. The number becomes useful when the record carries its definition with it.
Rolled throughput yield across process steps
Rolled throughput yield (RTY) measures how first-pass performance compounds through sequential process steps. VDS Automation’s FPY explanation shows the same step-by-step multiplication. Multiply the first-pass yield for each step:
RTY = FPY step 1 x FPY step 2 x FPY step 3
Suppose the three steps have first-pass yields of 90 percent, 95 percent, and 98 percent:
RTY = 0.90 x 0.95 x 0.98 = 0.8379, or 83.79 percent
This is not the same as final yield after rework. RTY asks whether the process passed cleanly at each step. Final yield asks how many units were accepted after the stated disposition process. A process can recover units through rework and still have a low RTY because the first-pass losses remain in the chain.
Worked example: one batch, three numbers
The following is an illustrative batch, not plant data or an industry benchmark. It turns the video segment’s 90 percent versus 98 percent contrast into a complete record.
| Batch field | Value | What it means |
|---|---|---|
| Units entering batch | 1,000 | The fixed denominator |
| First-pass-good units | 900 | Passed without rework |
| Units sent to rework | 80 | Failed the first attempt and entered another loop |
| Scrap units | 20 | Not accepted in the final disposition |
| Final-good units | 980 | 900 first-pass good plus 80 accepted after rework |
| First-pass yield | 90 percent | 900 / 1,000 x 100 |
| Final or blended yield | 98 percent | 980 / 1,000 x 100 |
The final yield is higher because the batch recovered 80 units through rework. That is useful information, but it should not erase the 80-unit first-pass failure. If the rework required inspection, machine time, cleaning, testing, or a constrained production slot, those loops belong in the follow-up record.
The RTY example above is separate. Its 95 percent middle step is an explicit illustrative assumption. Do not present 83.79 percent as the RTY of the batch table unless the batch actually has three measured process-step yields.
How to track FPY per batch
The transcript question is not only “what is the yield?” It is “what is the first-pass yield per batch?” That wording matters. A monthly average can hide which batches required a second or third attempt.
Start with a small record:
| Field | Record |
|---|---|
| Batch ID and date | The exact production batch and time window |
| Product, line, and shift | The context needed to compare like with like |
| Units entering | The fixed denominator |
| First-pass-good units | Units accepted without rework |
| Rework units | Units that needed another processing or inspection loop |
| Scrap units | Units not accepted in the final disposition |
| Final-good units | Accepted units after the stated disposition |
| FPY and final yield | Both percentages, with their formulas |
| Process step and reason | Where the first-pass loss appeared and what was observed |
| Owner and next action | Who can investigate or change the condition |
This is a diagnostic minimum, not a required software schema. The important control is that the denominator, rework count, and final disposition remain visible in the same record. The parent article’s practitioner research points to the same operating problem: teams can improve a local KPI while rework hours or lost production move into another report.
What to do when FPY is low and final yield is high
Treat the gap as a question, not as a verdict. Select one recurring batch, product, or line and trace the rework loop by defect, process step, time, and owner. Then ask whether the loop is avoidable, accepted by an old standard, or required by a current quality condition.
You do not need a new AI system to begin. You need a shared event identity and enough evidence for the next decision. Once the loss is defined, the Root Cause Analysis Workbook can help structure the investigation. Revisit the broader hidden manufacturing waste and rework cost framework when you need to price the labour, quality work, or capacity effect behind the percentage.
The practical rule is simple: use final yield to understand the final disposition, use FPY to expose first-pass loss, and use RTY to see how first-pass loss compounds across process steps.
