What is AQL (Acceptance Quality Limit)?

AQL is the quality-level value used to choose an acceptance sampling plan: instead of checking a whole lot, a random sample is drawn, and the lot is accepted or rejected according to the number of defective pieces found. In garments, the lot size (together with the inspection level) and the AQL value are cross-referenced in a table, which says how many pieces to inspect and how many defective pieces the sample may contain. A common misreading is that AQL 2.5 means “2.5% defects are allowed in the lot.” AQL is not the defect limit of a single lot; it is a value, set for the process average of successive lots, that selects the plan.

What AQL stands for

The abbreviation is expanded as “acceptance quality limit” in the title of ISO 2859-1 and as “acceptable quality level” in the US military standard MIL-STD-105E (1989); industry pages such as QIMA’s also use “acceptable quality limit.” In Turkish, the term is translated as kabul edilebilir kalite sınırı or seviyesi.

MIL-STD-105E defines AQL, for successive lots, as the quality level that is the limit of a satisfactory process average for sampling inspection (paragraph 3.1). According to the standard’s note, selecting an AQL means the plan will accept the great majority of lots as long as the average percent defective of the lots does not exceed that value. The same standard states three points explicitly:

• AQL is used together with the sample-size code letter to index plans (paragraph 4.4.1).
• Selecting or using an AQL gives the supplier no right to supply defective products (paragraph 4.4.2).
• AQL alone does not show the chance of accepting or rejecting a single lot; for relative risk, look at the plan’s operating characteristic (OC) curve (note to paragraph 3.1).

The standard: ISO 2859-1

ISO 2859-1 defines single, double and multiple sampling plans indexed by AQL for inspection by attributes. ISO 2859-1:2026 has been valid since January 22, 2026; the 1999 edition and its 2011 amendment were withdrawn on the same date (EVS, accessed October 11, 2026). State in the contract which edition you refer to. The US counterpart is ANSI/ASQ Z1.4 (QIMA; Wave PLM). MIL-STD-105E was cancelled in 1995; the definitions and table values in this entry are read from its text.

The plans are designed primarily for successive lots. If they are used for an isolated lot, look at the OC curves to find a plan that gives the protection you want (MIL-STD-105E, paragraph 1.2). ISO 2859-2:2020 defines a separate system for isolated lots, indexed by limiting quality (LQ) (EVS).

How an AQL plan is read

The code letter is found from the lot size and the inspection level. The code letter gives the sample size (n); the cell where the code-letter row meets the AQL column gives the acceptance number (Ac) and the rejection number (Re). At general level II, a lot of 1,201–3,200 pieces gets code letter K, a sample of 125 pieces: at AQL 2.5, accept on 7 and reject on 8; at AQL 4.0, accept on 10 and reject on 11 (MIL-STD-105E, Tables I and II-A; Wave PLM gives the same 125-piece sample). Under normal inspection the lot passes if the number of defective pieces is equal to or below the acceptance number, and fails if it is equal to or above the rejection number. Confirm the values against the ISO 2859-1:2026 tables before writing them into a contract. The full table, a worked example and the counting method are in AQL inspection for garments.

Common values in garments

The AQL value can be chosen separately for each defect group (MIL-STD-105E, paragraph 4.4.3). The same standard defines critical, major and minor defects in paragraphs 3.5, 3.14 and 3.16 (our paraphrase below); the examples are from Wave PLM.

Wave PLM (May 22, 2026) reports these three values and general inspection level II as the common setting in apparel; QIMA also writes that 0 for critical defects and 2.5 are common for general consumer goods. These are industry practice, not part of the standard; what binds the factory is the value the buyer writes into the purchase order. According to QIMA, a single critical defect stops the shipment (September 27, 2022). QIMA’s pre-shipment inspection (PSI) is carried out when at least 80% of the order is finished and packed, and the major and minor findings are compared with the AQL table at that inspection. For where inspection sits in the production flow, see garment manufacturing process; for providers, see quality control and inspection services.

How AQL differs from related terms

• Inspection level: Together with AQL, it determines the plan. The level sets the relationship between lot size and sample size (paragraph 4.9.1); AQL sets the acceptance and rejection numbers for the sample.
• OC curve (operating characteristic curve): Shows the probability that a plan accepts a lot at a given true defect rate. AQL is a single value; the curve shows the chance of acceptance.
• LQ (limiting quality): The plan index used for isolated lots in ISO 2859-2.
• Sampling: The act of drawing the sample. AQL is not the sampling itself but the value that selects the plan.
• Laboratory testing: Performance properties such as shrinkage, colorfastness and chemical compliance are verified by laboratory testing, not by AQL inspection on site (Tekstil Rehberi note); see textile testing laboratories.

Frequently asked questions

Does AQL 2.5 mean that 2.5% defects are allowed in a lot?

No. AQL is an index that selects the plan (paragraph 4.4.1) and does not show the chance of acceptance of a single lot (note to paragraph 3.1). By Tekstil Rehberi’s calculation (random sample, binomial distribution): for a lot of 2,000 pieces inspected with plan K (125 pieces, major AQL 2.5, accept on 7), a lot whose true major defect rate is 5% is accepted with a probability of about 71%; at 8% the probability falls to about 21%. The full table is in the AQL inspection for garments guide.

Does AQL 0 for critical defects guarantee a defect-free lot?

No. A plan with an acceptance number of 0 stops the lot if a single critical defect is found in the sample, but the sample is only a small part of the lot. By Tekstil Rehberi’s calculation, if 1% of a 2,000-piece lot (20 pieces) has a critical defect, the probability that none shows up in a random sample of 125 pieces is about 27% (28% with the binomial approximation). Do not leave critical risks to sampling alone; manage them also with 100% inspection, process controls and laboratory testing.

What AQL is normal for garments?

According to industry sources, the common setting is 0 for critical, 2.5 for major and 4.0 for minor defects, with general inspection level II (Wave PLM, May 22, 2026; QIMA for critical 0 and 2.5). This is practice, not a standard: the value that binds the factory is the one written into the purchase order.