Aql
Does AQL mean the lot quality is at the AQL percentage?
No. AQL is the worst tolerable process average or acceptance threshold used for sampling decisions, not a direct measured defect rate for every lot. It is a statistical policy parameter, not a product quality score.
What do “accept number” and “reject number” mean?
They are the counts from the AQL table that tell inspectors the maximum number of defects allowed in the sample before lot rejection. Sample size and accept/reject numbers come from the lot size, inspection level, and AQL under ISO 2859-1/ANSI/ASQ Z1.4.
Can AQL be used for critical, major, and minor defects separately?
Yes. Industry practice commonly assigns different AQL limits by defect severity, so safety-related critical defects can have much tighter acceptance criteria than cosmetic minor defects. This lets a receiving process balance safety risk with commercial practicality.
Acceptance Quality Limit (AQL) is the worst tolerable process average in acceptance sampling. It is a statistical cutoff used in manufacturing and supply-chain inspection to decide whether a lot is accepted or rejected based on a random sample. Under ISO 2859-1/ANSI/ASQ Z1.4, AQL specifies how many defects are allowed in the sample before the entire batch is rejected.
On the shop floor, AQL is used at incoming inspection, in-process inspection, and final lot release for purchased parts, subassemblies, and finished goods. For raw material tracking, receiving teams use a statistically valid sample to decide whether coils, castings, fasteners, molded parts, or packaged components can be released to inventory instead of 100% inspection. In warehousing, AQL drives receiving QA gates: operators sample cartons or pallets, record defect counts by severity, and either put the lot away, send it to quarantine, or escalate to supplier corrective action when the accept number is exceeded. In procurement, AQL is written into supplier quality agreements, defining acceptable defect rates for critical, major, and minor defects so buyers and suppliers share the same rejection rule before shipment leaves the dock. Operationally, this matters when volume is high and full inspection would bottleneck receiving; the sampling plan cuts inspection workload while enforcing a quality threshold.
Misreading AQL as zero-defect acceptance: Teams assume a lot with one sample failure is automatically unusable, but AQL is a defined sampling threshold, not a promise of perfection. This causes unnecessary rejection of acceptable lots and wasted quality time.
Using the wrong defect class or sampling level: Applying the same AQL to critical safety defects and cosmetic blemishes can over-reject good material or under-control serious nonconformances. Separate AQL values by severity to align sampling stringency with risk.
Letting sampling rules mismatch real lot behavior: Mixed, repacked, or poorly traceable lots can hide localized defect clusters from random samples, releasing bad material into production and creating downstream scrap, line stoppages, or rework at the assembly cell.