AQL Reality Check

Every AQL calculator tells you the sample size. This one tells you what the plan costs you — in dollars.

Every value above is a starting default, and yours to change. The intangible multipliers are your own assumption about what a bad lot really costs you — they are not an industry statistic.

Your sampling plan — ISO 2859-1 / ANSI Z1.4 · Level II · Normal

Code L Sample 200 Accept 10 Reject 11

That’s where every other calculator stops. Below is what the plan actually does with your money.

01 · The discount you never claim
$13,824 / yr

Suppliers with a clean run of accepted lots may be moved to reduced inspection — sample 80 instead of 200. If 50% of your 144 yearly inspections qualified, the standard offers you roughly $13,824 of your $46,080 annual inspection spend back.

A potential saving, not an entitlement: reduced inspection is permitted by the responsible authority once the switching score reaches 30 and production is steady (clause 9.3.3). Almost no importer claims it, because claiming it requires lot history per supplier — a series almost nobody keeps.

Why this happens → Paying for a System, Buying a Gate
02 · What slips through — priced
$1,600$1,600 / lot

A lot at 4.0% defective — 60% worse than your contracted 2.5% — still passes this plan 82% of the time. When it does, ≈200 defective garments ship in this one lot. The range runs from $1,600 in direct cost up to $1,600 at your chosen ×1 assumption.

Reliable rejection (9 times in 10) only begins near 7.6% defective — 3.0× your contract. And the costliest line has no number at all: if a lot like this costs you the customer, what is their next PO worth?

Why this happens → When AQL Inspection Fails Your Brand
03 · The price of a second spin
$32075%

A rejected lot at this plan’s 50/50 quality level (5.3%) passes an unchanged re-inspection 50% of the time. Book two attempts and the odds reach 75% — rework or no rework. Each attempt bills $320.

Clause 7.6 of ISO 2859-1 gives the buyer the fix: require the factory’s 100% re-examination record, and designate tightened severity for the resubmission. Qualityiris recommends that severity — the inspector decides, and an override is recorded with its reason.

The most common AQL myth: AQL 2.5 does not mean your lot is guaranteed ≤2.5% defective. It means a lot at 2.5% would usually be accepted. The two are not the same — the difference is everything above.

Isolated lot — new factory, spot buy, single program? The standard family has a different plan for that: ISO 2859-2, indexed by the limiting quality you refuse to accept. Its plans hold the probability of accepting a lot at that level to no more than about 10% — the consumer’s risk, where ISO 2859-1 fixes the producer’s risk instead. An LQ value is not equivalent to an AQL value. QualityIris applies both standards — and knows which one your lot needs.