AQL sampling plan calculator (ANSI/ASQ Z1.4)
Enter your lot size, inspection level and AQL; get the sample size and accept/reject numbers from the published single sampling tables. Free, runs in your browser.
What AQL actually means
AQL is the acceptable quality limit: the worst average nonconforming level that a sampling scheme will still routinely accept across a series of lots. It describes the behaviour of the scheme over time. It is not a target, a tolerance, or a statement about the lot in front of you.
A lot accepted at AQL 2.5 has not been shown to be free of defects. It has only failed to produce enough nonconforming pieces in the sample to reach the reject number. A rejected lot has not been proven bad either. Small samples carry real risk in both directions, and both risks are published: the operating characteristic curve for each plan shows the probability of acceptance at any true defect rate.
Tightening the AQL or moving to a larger sample changes the shape of that curve. It never removes the risk, and no sampling plan can turn a sample result into a statement about every piece in the lot. An accepted lot does not tell you the true defect rate of the lot, whether the process that made it is stable or capable, or whether the defects that are present matter to the customer who receives them.
How to choose an inspection level
The level decides how much of the lot you look at. It is a business decision about how much inspection you are willing to pay for, made before you pick an AQL. If a drawing, a control plan or a purchase order names a level, that one wins.
The sample size code letter table
Find the row for your lot size, read across to your inspection level and take the letter. That letter plus the AQL and the severity picks the row in the single sampling table, which is where the sample size and the accept and reject numbers come from.
| Lot size | S-1 | S-2 | S-3 | S-4 | I | II | III |
|---|---|---|---|---|---|---|---|
| 2 to 8 | A | A | A | A | A | A | B |
| 9 to 15 | A | A | A | A | A | B | C |
| 16 to 25 | A | A | B | B | B | C | D |
| 26 to 50 | A | B | B | C | C | D | E |
| 51 to 90 | B | B | C | C | C | E | F |
| 91 to 150 | B | B | C | D | D | F | G |
| 151 to 280 | B | C | D | E | E | G | H |
| 281 to 500 | B | C | D | E | F | H | J |
| 501 to 1,200 | C | C | E | F | G | J | K |
| 1,201 to 3,200 | C | D | E | G | H | K | L |
| 3,201 to 10,000 | C | D | F | G | J | L | M |
| 10,001 to 35,000 | C | D | F | H | K | M | N |
| 35,001 to 150,000 | D | E | G | J | L | N | P |
| 150,001 to 500,000 | D | E | G | J | M | P | Q |
| 500,001 and over | D | E | H | K | N | Q | R |
Run the calculator above and the cell it used is marked here. On a phone, swipe the table sideways for the special levels.
Normal, tightened and reduced: the switching rules
Z1.4 is a scheme, not a single plan. Severity is meant to move with the evidence: a supplier who keeps failing gets looked at harder, and a supplier with a long clean record can be looked at less. The rules are mechanical, and they depend on your lot history, which is exactly the part a one-shot calculator cannot see.
Every new product, tool or supplier begins here, and every return from tightened or reduced lands here.
Switch when 2 of 5 consecutive lots are rejected on original inspection. Return to normal after 5 consecutive lots are accepted. If 5 lots are rejected while on tightened, the standard says to discontinue inspection under Z1.4 until the process is fixed.
Allowed when the last 10 lots were accepted on original inspection, the total nonconforming is under the Table VIII limit, production is steady and the responsible authority agrees.
Under reduced inspection the accept and reject numbers are not consecutive. If the count lands in the gap between them, you accept the lot and go back to normal inspection for the next one. This calculator returns the plan for whichever severity you select; it does not track your history and it does not switch severity for you.
Z1.4, MIL-STD-105E and ISO 2859-1
MIL-STD-105E was the United States military standard for attribute sampling until it was withdrawn in 1995. ANSI/ASQ Z1.4 carries the same tables forward as a civilian standard. ISO 2859-1 is the international line of the same family, with the same code letters and the same single sampling tables.
Drawings and purchase orders in the field still name all three, often for the same plan. For the general levels and the common AQLs, the sample size and the accept and reject numbers line up. The differences show up around the edges: the wording on switching, the treatment of limiting quality, and how each document handles reduced inspection. Read the clause in the contract before you treat the three names as interchangeable.
This calculator covers SINGLE sampling only. Double and multiple sampling plans, which let you take a second or later sample before deciding, are not included. The AQL columns above 10, expressed as defects per 100 units rather than percent nonconforming, are out of scope too.