AQL Sampling Plan

Lot size and inspection level (general I–III, special S-1 to S-4) give a sample-size code letter from Table I of ANSI/ASQ Z1.4 (identical in ISO 2859-1). The letter fixes the sample size n; the letter and the AQL fix the accept number Ac and reject number Re = Ac + 1 from the single-sampling, normal-inspection table (II-A). Where that table shows an arrow instead of a plan, the standard says to use the first plan below (a larger sample) or above it — the page walks the arrow and tells you it did. Note the table's shape: after the Ac 0 cell of a row come an up arrow and a down arrow before Ac 1, so a slightly looser AQL can send you to a smaller sample and a slightly tighter one to a larger sample. If n reaches the lot size, the whole lot is inspected.

AQL is the worst process average the plan is designed to accept most of the time — it is not a guarantee about any single lot. The operating characteristic shows the actual probability of acceptance for each lot quality, computed with the binomial distribution (exact for large lots, close for small ones). The page reports Pa at the AQL, Pa at twice the AQL, and the limiting quality where Pa falls to 10 % — the quality the plan will almost always reject.

Single sampling, normal inspection, AQL 0.010–10 % (percent nonconforming). Tightened and reduced inspection, double and multiple plans, and the switching rules are not included. The tables in this utility were transcribed cell by cell from MIL-STD-105E (1989), a public-domain US Department of Defense standard whose Tables I and II-A are the same plans as ANSI/ASQ Z1.4 and ISO 2859-1; the dataset records the pages used. Inputs stay in your browser; the same four anonymous usage counts as the rest of the site apply.