Determine inspection sample size and accept/reject numbers based on ANSI/ASQ Z1.4 AQL tables. Plan incoming and in-process inspections.
Acceptance Quality Level (AQL) sampling determines how many units from a production lot must be inspected and how many defects are acceptable before the lot is accepted or rejected. The ANSI/ASQ Z1.4 standard (also known as MIL-STD-1916 successor and ISO 2859-1) provides standardized tables that map lot sizes and AQL levels to sample sizes and accept/reject numbers.
AQL sampling balances inspection cost against quality risk. Inspecting every unit (100% inspection) is expensive and often impractical. AQL plans provide statistically justified sample sizes that control the probability of incorrectly accepting bad lots or rejecting good lots.
This calculator takes your lot size, AQL level, and inspection level to determine the sample size and accept/reject criteria, simplifying the lookup process that quality inspectors traditionally perform with printed tables.
This measurement forms a critical foundation for capacity planning, helping teams align production capabilities with demand forecasts and strategic business objectives throughout the planning cycle.
AQL sampling provides a standardized, statistically sound method for lot acceptance that is recognized worldwide. It eliminates subjective sample size decisions and ensures consistent quality decisions across suppliers, inspectors, and facilities. Consistent measurement creates a reliable baseline for tracking improvements over time and demonstrating return on investment for process optimization initiatives.
Based on ANSI/ASQ Z1.4 tables: 1. Lot size → Sample size code letter (using inspection level table) 2. Code letter + AQL → Sample size (n), Accept number (Ac), Reject number (Re) If defects found ≤ Ac → Accept lot If defects found ≥ Re → Reject lot
Result: Sample size = 200, Ac = 5, Re = 6
For a lot of 5,000 units at AQL 1.0% with General Inspection Level II, the code letter is L, giving a sample of 200 units. Accept the lot if 5 or fewer defects are found; reject if 6 or more.
The Z1.4 tables are organized in two stages. First, a lot-size/inspection-level table converts your lot size and inspection level to a code letter (A through R). Second, a code-letter/AQL table converts the code letter and desired AQL to sample size (n), accept number (Ac), and reject number (Re).
For critical safety defects, AQL sampling may not provide adequate protection. Consider 100% inspection (with adequate gage capability) for characteristics where even one defect poses serious risk to consumers or end users.
AQL sampling is the international standard for import/export quality inspection. Third-party inspection companies use AQL plans to evaluate shipments on behalf of buyers, providing an objective quality gate at the point of manufacture.
AQL stands for Acceptance Quality Level (or Limit). It is the maximum percentage of defective units in a lot that is considered acceptable as a process average. An AQL of 1.0% means lots averaging 1% defective will be accepted most of the time.
General Level II is the default and most commonly used. Level I uses smaller samples (less protection), Level III uses larger samples (more protection). Special levels S-1 to S-4 are for very small samples or destructive testing.
Rejected lots may be 100% sorted (screening), returned to the supplier, reworked, or scrapped. The disposition depends on your quality agreement and the severity of defects found.
No. AQL sampling controls the long-run average quality of accepted lots. Individual lots with higher defect rates can be occasionally accepted (producer's risk) and good lots can be occasionally rejected (consumer's risk).
Per Z1.4, you switch to tightened inspection after 2 of 5 lots are rejected, back to normal after 5 consecutive acceptances, and to reduced after 10 consecutive normal acceptances with low defect counts.
Yes. AQL sampling can be applied to any countable population — invoices, transactions, data records, etc. Define a "unit" and a "defect" appropriately for your service context.