Why cost per part produced rather than per defective part?

Because that is the unit everything else uses. Cost per defective part is useful for understanding an event, but it cannot be added to a per-part cost model or compared with the cost of curing. Spreading the event across the parts produced puts quality cost on the same line as material, labour and energy, which is where a decision actually gets made.

Why was the customer-impact input removed?

Because it was a lump with no derivation, and a placeholder is worse than an omission. It makes a total look complete while being invented, and everything downstream inherits the invention. Customer impact is either unknown, in which case say so, or it is a specific known consequence, in which case it belongs in the disposition figure with a note recording what it is.

Why is a small defect event so expensive per part?

Because containment does not scale with the event. Sorting, quarantining, paperwork and the engineering time to find the cause are much the same for twenty parts as for two hundred, so a small event spreads a fixed cost over few parts. The practical consequence is that on a line dominated by containment, the number of events matters far more than the number of parts in each.

How should I value a scrapped part?

At its full accumulated value at the cure station. Material, all prior operations and the labour in them, not at the cost of the coating. Everything spent on the part up to that point is lost with it, which is why a cure defect late in a process is so much worse than the same defect early, and why the scrap case costs several times the rework case in the examples here.

What about parts reworked more than once?

Count the part once and charge the per-part cost for each pass, either by raising the per-part figure or by entering the passes as separate events. Counting the part twice inflates the defect rate and makes the PPM row wrong, which matters if that figure feeds a quality report.

Does this cover defects that reach a customer?

No, and the omission is significant. A defect found in-house is the cheap outcome; one that ships carries returns, investigation, replacement, freight and reputational cost that routinely dwarf everything on this page. Undercure is exactly the failure mode most likely to escape. A part soft beneath a cured skin looks finished, so the true quality cost of poor cure control is systematically higher than any event this page prices.

UV Curing calculator

UV Rework Cost Calculator: Quality Cost per Part Produced

Price a cure defect event and put the answer in a unit that composes. Enter the defective parts, the cost of dealing with each, the disposition and containment labour, and the total produced in the period. The calculator returns the quality cost carried by every part produced, the event total behind it, the cost per defective part, the material share, and the defect rate the event represents. Cost per part produced is the headline because it is the only form in which quality cost can be set beside the cost of curing, and that comparison is usually the point.

What this calculator does

Price a cure defect event and express it as the quality cost carried by every part produced, so it can be compared with the cost of curing them.
Use it for putting a quality cost into a per-part cost model, comparing quality cost against cure cost on the same basis, showing why small containment events are disproportionately expensive, building the case for dose monitoring from avoided events, pricing the difference between a reworkable and a scrapped defect.
Price a cure defect event and express it as the quality cost carried by every part produced, so it can be compared with the cost of curing them.

Formula used

Rework / scrap material = defective parts × cost per part
Total cure defect event cost = material + disposition labour
Quality cost per part produced = event cost ÷ total parts produced
Cost per defective part = event cost ÷ defective parts
Defect rate = defective parts ÷ total parts produced × 1,000,000

Inputs explained

Defective parts: Parts in this event requiring rework or scrapping for cure reasons.
Rework or scrap cost per part: Cost of dealing with one defective part: stripping and re-coating, or the full value lost if it is scrapped. Use the higher of the two where both happen.
Disposition / containment labour: Sorting, inspecting, quarantining, paperwork and any line stoppage. The fixed cost of the event regardless of how many parts it contains.
Total parts produced in the period: All parts made in the period this event belongs to. Converts an event cost into the quality cost every part carries.

How to use the result

Best suited to putting a quality cost into a per-part cost model, comparing quality cost against cure cost on the same basis, showing why small containment events are disproportionately expensive, building the case for dose monitoring from avoided events, pricing the difference between a reworkable and a scrapped defect.
Prices one event. A period's total quality cost is the sum of its events, and the fixed containment cost recurs with each one. Excludes customer impact entirely, which is deliberate: a placeholder for it is worse than its absence, because it makes a total look complete while being invented. Ignores the schedule effect of a containment. Parts held while a decision is made can miss a shipment, and that cost is real and situational. Says nothing about defects that escaped. A part that reached a customer costs a multiple of one caught in-house, and it does not appear in any event this page prices. Assumes uniform per-part cost, when a mixed event of reworkable and scrapped parts has two very different figures.

Common questions

Why cost per part produced rather than per defective part? Because that is the unit everything else uses. Cost per defective part is useful for understanding an event, but it cannot be added to a per-part cost model or compared with the cost of curing. Spreading the event across the parts produced puts quality cost on the same line as material, labour and energy, which is where a decision actually gets made.
Why was the customer-impact input removed? Because it was a lump with no derivation, and a placeholder is worse than an omission. It makes a total look complete while being invented, and everything downstream inherits the invention. Customer impact is either unknown, in which case say so, or it is a specific known consequence, in which case it belongs in the disposition figure with a note recording what it is.
Why is a small defect event so expensive per part? Because containment does not scale with the event. Sorting, quarantining, paperwork and the engineering time to find the cause are much the same for twenty parts as for two hundred, so a small event spreads a fixed cost over few parts. The practical consequence is that on a line dominated by containment, the number of events matters far more than the number of parts in each.
How should I value a scrapped part? At its full accumulated value at the cure station. Material, all prior operations and the labour in them, not at the cost of the coating. Everything spent on the part up to that point is lost with it, which is why a cure defect late in a process is so much worse than the same defect early, and why the scrap case costs several times the rework case in the examples here.
What about parts reworked more than once? Count the part once and charge the per-part cost for each pass, either by raising the per-part figure or by entering the passes as separate events. Counting the part twice inflates the defect rate and makes the PPM row wrong, which matters if that figure feeds a quality report.

Last reviewed 2026-08-25.