The true cost of rework is the incremental cost of making a nonconforming unit acceptable again, not simply the material in the hold area. Start with added material, rework labour, QA or retest, processing and machine time, and the cleaning, maintenance, or documentation work the event creates. Add capacity opportunity cost when the rework displaces feasible production.
This page gives you a formula, a rework-or-scrap break-even test, and an expected-cost calculation for second and third attempts. The numbers in the worked examples are illustrative, not plant data or industry benchmarks.
What counts as rework in manufacturing
For this article, rework is additional processing intended to bring a nonconforming unit or batch back to the original specification. Scrap is a different disposition: the material or product will not return to the intended saleable output. The terms should stay separate in the event record.
The American Society for Quality classifies waste, scrap, rework or rectification, and failure analysis as internal failure costs within the cost-of-quality framework.
Keep first-pass yield separate as well. First-pass yield measures output that met requirements without a later rework pass. A unit can be acceptable at the end and still be a first-pass failure if it needed correction or retesting. SAP’s manufacturing guidance separates first-pass output from later-pass or retest output.
The rework process and where cost enters it
The cost enters in stages. If the event is tracked only as “rework completed,” the plant loses the evidence needed to decide whether the next attempt is worthwhile.
| Stage | Record to check | Cost that can enter |
|---|---|---|
| Detect and hold | Nonconformance, batch or work order, quantity | Quarantine handling, movement, waiting, and disposition time |
| Decide the route | Disposition decision and approved instructions | Quality review, engineering time, and documentation |
| Perform the correction | Labour entries, added materials, machine or utility time | Direct rework labour, replacement material, processing, and machine use |
| Verify and release | Retest, inspection, cleaning, release record | QA or lab time, cleaning, maintenance, compliance documentation, and release delay |
| Close the event | Capacity state, owner, next action, final disposition | Displaced work, overtime, schedule disruption, or a repeat investigation |
The supplied interview names the same wider boundary: material, labour, QA testing, utilities, capacity, maintenance, cleaning, and compliance documentation. Treat that list as a scoping checklist, not as a universal accounting formula.
How to calculate rework cost
Use this as the first-pass operational formula:
Incremental rework cost = added material + rework labour + QA or retest + processing and machine time + cleaning, maintenance, and documentation
Then calculate the decision cost:
Decision cost = incremental rework cost + capacity opportunity cost
A calculator can give you a useful floor from labour, added material, and reinspection. The original manufacturing cost already spent before the disposition decision is not added again as incremental rework cost. CalcBee’s rework calculator makes the same distinction in its explanation of incremental cost.
Capacity opportunity cost is conditional. If the line has genuine spare capacity during the event window, record the opportunity-cost field as zero and keep the direct costs. If the line is constrained and rework displaces a feasible product, use the plant’s approved contribution-margin basis for the displaced work. Do not use revenue as a substitute unless finance has approved that basis.
The formula is a decision boundary, not a GAAP method, standard cost policy, legal requirement, or food-safety rule. Use the plant’s finance and quality definitions when you move from an illustrative calculation to a live disposition.
Rework or scrap: the break-even test
The useful comparison is not “rework cost versus the original cost of the part.” Much of that original cost is already sunk. Compare the avoidable decision cost with the value preserved by choosing rework instead of scrap.
Value preserved by rework = (recoverable value per unit – net scrap recovery per unit) x units
Choose rework when:
incremental rework cost + capacity opportunity cost < value preserved over net scrap recovery
Choose scrap, or another approved disposition, when the inequality reverses or when the quality risk is not acceptable. The quality decision is a gate before the economic comparison.
Illustrative two-scenario example
Assume 100 held units, a recoverable value of $60 per unit if reworked and a net scrap recovery of $12 per unit. The value preserved by rework is (60 - 12) x 100 = $4,800.
| Cost or value field | Line with spare capacity | Constrained line |
|---|---|---|
| Added material | $300 | $300 |
| Rework labour | $900 | $900 |
| QA and retest | $250 | $250 |
| Utilities and machine time | $350 | $350 |
| Cleaning and handling | $150 | $150 |
| Incremental rework cost | $1,950 | $1,950 |
| Capacity opportunity cost | $0 | $3,600 |
| Decision cost | $1,950 | $5,550 |
| Value preserved over net scrap recovery | $4,800 | $4,800 |
| Economic result in this illustration | Rework, $2,850 below preserved value | Scrap, $750 below rework value |
The same held units produce a different economic answer because the line state changed. The $3,600 capacity value is an illustrative contribution-margin assumption, not lost revenue or a plant benchmark. Replace it with the approved basis for the event window.
Rework loops: why the second attempt costs more
A second attempt is not free just because it uses the same routing. Price the probability of another attempt.
If the cost per attempt is stable, use:
Expected rework cost = C1 + (p2 x C2) + (p2 x p3 x C3)
Where C1, C2, and C3 are the costs of the first, second, and third attempts, p2 is the probability that the first attempt needs a second attempt, and p3 is the conditional probability that the second attempt needs a third.
For an illustrative unit with $19.50 per attempt, a 20 percent conditional second-attempt probability, and a 30 percent conditional third-attempt probability:
$19.50 + (0.20 x $19.50) + (0.20 x 0.30 x $19.50) = $24.57 expected cost per unit
The expected cost is 1.26 times the single-attempt cost. That multiplier is not a benchmark. It is a transparent way to prevent a repeat loop from disappearing into separate work orders. If the second or third attempt uses different labour, material, or QA, use a different C2 or C3.
What to record for each rework event
Give the event one identity that production, quality, maintenance, planning, and finance can append to. A minimum record includes:
| Field | Why it matters |
|---|---|
| Event ID, batch or work order | Keeps all records tied to one event |
| Product, line, process step, and quantity | Sets the comparison boundary |
| First-pass result and final disposition | Keeps rework visible even when final output passes |
| Added material and direct rework labour | Starts the incremental cost |
| QA, retest, utilities, machine, cleaning, and documentation time | Captures the work outside the rework order |
| Capacity state and approved valuation basis | Shows whether the line was constrained |
| Attempt number and prior attempt result | Enables the loop calculation |
| Owner, suspected cause, next action, and verification date | Moves the record into root-cause work |
Unknown values stay unknown. They are evidence gaps to resolve, not zeros to make the calculation look complete.
Use the calculation to choose the next investigation
The number is useful only if it changes a decision. Start with one repeatable event, verify the cost fields, and use the attempt history and capacity state to decide whether the next action is a process correction, a standard review, a supplier investigation, or a controlled technology assessment.
The Root Cause Analysis Workbook is the practical next step for turning the event record into a structured investigation. For the wider plant-level cost boundary, see how to measure the hidden cost of manufacturing waste and rework. If the event also changed first-pass performance, compare it with First Pass Yield vs Final Yield.
This page is an English video-spoke guide. Its examples are illustrative and are not plant data, a survey, an industry benchmark, or a universal accounting formula. It does not provide legal, regulatory, food-safety, GAAP, or finance advice.
