Paste this where you want the calculator to appear. It works on any site — WordPress, Squarespace, Webflow, Ghost or plain HTML — and needs no JavaScript of yours. It carries a link back here, which is the only thing we ask for it.
The accept number is not a percentage of the sample. It comes from the standard table.
Sample Size
—pcs
Units to inspect at General Level II
Accept / Reject Decision
Accept on or Below
—defects
Reject on or Above
—defects
Defects Still Allowed
—no.
Observed Defect Rate
—%
This is normal inspection at General Level II. Switch to tightened inspection after two rejected lots in five, as the standard requires.
Using this calculator
About the AQL Sampling Plan Generator (ISO 2859-1)
The formula
This is the expression the tool evaluates. Every term is named underneath, with the unit it must be supplied in.
Sample size is looked up, not calculatedlotSize -> code letter -> sampleSize (ISO 2859-1 Table 1, General Inspection Level II)
There is no formula. Lot size maps to a code letter and the code letter maps to a sample size, and the steps are deliberately coarse: every lot from 3,201 to 10,000 pieces gets the same sample of 200.
Also table values. The accept number is not a percentage of the sample and cannot be derived from the AQL by multiplication — at AQL 2.5 on a sample of 200 it is 10, where 2.5% of 200 would suggest 5.
The decision, and the observed ratemarginToAccept = acceptNumber - defectsFound defectRate = defectsFound / sampleSize x 100
A margin of zero or above accepts the lot. The observed defect rate is reported for information and plays no part in the decision — the accept number does.
Symbols used above
Symbol
Stands for
Unit
AQL
Acceptance Quality Limit — the worst average quality still considered acceptable as a process average, not a per-lot maximum
%
Ac
Accept number — the most defects that may be found and the lot still pass
defects
Re
Reject number — always Ac plus one in a single sampling plan
defects
n
Sample size, drawn at random from across the whole lot
pcs
How the result is derived
Step by step, from the values you type to the figure on screen.
The lot size selects a sample size code letter from the General Inspection Level II column, which is the default level the standard prescribes when no other is specified.
The code letter gives the sample size — the number of units to draw at random from across the lot, not from one carton or one production hour.
The code letter and the chosen AQL together give the accept number from the single sampling table for normal inspection.
The reject number is the accept number plus one, so there is no ambiguous middle outcome: every lot either passes or fails.
The defects found are compared against the accept number, and the remaining allowance is reported so an inspection still in progress knows where it stands.
What each input means
Where to read each value on the floor, the unit it must be in, and the range the tool accepts.
Input
Unit
Accepted range
Default
What it means
Lot Size
pcs
2 to 10000000 pcs
5000
Acceptance Quality Limit
—
AQL 1.0 — critical / premium · AQL 1.5 — major defects, strict · AQL 2.5 — major defects, standard · AQL 4.0 — minor defects
2.5
Defects Found in Sample
no.
0 to 100000 no.
8
What the tool returns
The headline figure and every supporting value it is built from.
Output
Unit
What it tells you
Sample Size (headline result)
pcs
Units to inspect at General Level II
Accept on or Below
defects
Reject on or Above
defects
Defects Still Allowed
no.
Observed Defect Rate
%
Worked example
Given
Lot size
5,000 pcs
AQL
2.5 (major defects, standard)
Defects found
8
Substituting
lot 5,000 falls in the 3,201-10,000 band -> code letter L -> sample 200code L at AQL 2.5 -> Ac = 10, Re = 11margin = 10 - 8 = 2 defects still allowedrate = 8 / 200 x 100 = 4.00%decision = 8 <= 10, so the lot is ACCEPTED
Answer
Sample size
200 pcs
Accept on or below
10 defects
Reject on or above
11 defects
Defects still allowed
2
Observed defect rate
4.00%
The lot passes with an observed defect rate of 4.00% against an AQL of 2.5 — which looks contradictory and is not. The AQL describes an acceptable process average over many lots; the accept number is set so that a process genuinely running at 2.5% passes about 95% of the time, which means it must tolerate individual lots that sample worse. Rejecting this lot because 4.00% exceeds 2.5% would be applying the standard incorrectly, and it is the most common way this table is misread.
How to use it
Work through the input groups in order — Lot and Inspection Result. The defaults are a realistic case, so you can change one value at a time and watch what moves.
There is no calculate button. Every figure recalculates as you type or drag, which is what makes this usable for a what-if sweep rather than a single answer.
Read Sample Size in the dark results panel — that is the headline figure, expressed in pcs.
Check the supporting rows underneath (Accept on or Below, Reject on or Above, Defects Still Allowed and Observed Defect Rate) before acting on the headline — they are where an implausible input usually shows itself first.
Reset to defaults returns every field to the reference case, which is the quickest way to check whether a surprising result came from the tool or from an input you had changed earlier.
Where this is used
Final random inspection of a finished garment lot before shipment, which is where this plan is used most.
Incoming inspection of fabric rolls, trims and accessories against an agreed AQL.
Writing an inspection clause into a purchase order — the AQL, the inspection level and the sampling standard together are the clause.
Resolving a dispute about whether a lot should have passed, by reconstructing the plan that applied at the time.
Reading the result
Typical bands and what each one is telling you.
Value
What it indicates
AQL 1.0
Critical characteristics and premium brands. Roughly halves the accept number against 2.5 on the same sample.
AQL 1.5
Major defects on a strict programme, common for branded apparel.
AQL 2.5
Major defects, the general apparel default and the figure most buying offices assume when none is stated.
AQL 4.0
Minor defects. Usually quoted alongside 2.5 for major in the familiar 2.5/4.0 pairing.
Margin of 0
The lot passes and there is nothing left. One more defect anywhere in the remaining sample rejects it.
Margin below 0
Rejected. Under the standard the lot is returned for sorting or rework and re-offered, not renegotiated defect by defect.
Assumptions and limits
Single sampling, normal inspection, General Inspection Level II. Double and multiple sampling plans exist in the same standard and give different numbers for the same lot.
Normal inspection only. The standard requires switching to tightened inspection when two of five consecutive lots are rejected, and this tool does not track that history — it is a per-lot calculator and switching is a programme decision.
The sample is drawn at random from the whole lot. A sample taken from whatever was nearest the door is not a sample, and no table can repair it.
One AQL at a time. Real inspections usually run separate AQLs for major and minor defects, which means running this twice and passing both.
Defects are counted, not defectives. Classifying a garment with three faults as three defects or as one defective unit changes the answer, and the choice must be agreed before the inspection.
Standards and further reading
ISO 2859-1 — sampling procedures for inspection by attributes, indexed by acceptance quality limit. The source of every number this tool returns.
ANSI/ASQ Z1.4 — the American equivalent, with the same sample sizes and accept numbers for the plans used here.
ISO 2859-2 — sampling by limiting quality, for isolated lots where no process average exists to describe.
Questions people ask
Why is the accept number 10 and not 2.5% of the sample?
Because the AQL is not a defect budget for one lot. The accept numbers are chosen so that a supplier whose process genuinely averages the AQL passes about 95% of lots — anything stricter would reject good suppliers on ordinary sampling variation. On a sample of 200 at AQL 2.5, that operating characteristic puts the accept number at 10. Reading the AQL as a per-lot ceiling and computing 2.5% of 200 gives 5, and would fail roughly half the lots from a supplier who is meeting the specification.
The lot is 5,000 pieces and the sample is only 200. Is 4% of the lot enough?
Yes, and the ratio falls further as lots grow — 1,250 units from a lot of a million is 0.125%. Statistical confidence depends on the absolute size of the sample far more than on the fraction of the lot it represents, which is why the code letters step so coarsely. The same reasoning is why a 3,300-piece lot and a 9,900-piece lot share a sample of 200.
What happens when a lot is rejected?
Under the standard the lot goes back for 100% sorting or rework and is then re-offered, and the standard also requires switching to tightened inspection once two of any five consecutive lots have been rejected. What should not happen is re-inspecting the same lot with a fresh sample hoping for a pass, which discards the statistical basis of the plan entirely.
