Selecting AQL Levels And Sample Sizes For Production

Acceptance Quality Limit inspection, a sampling method used to decide whether a lot meets quality standards.

AQL Inspection Acceptance Quality Limit inspection, a sampling method used to decide whether a lot meets quality standards. Choosing the correct AQL level and sample size requires balancing acceptable defect risk, inspection cost, and product criticality; this article explains the decision points and provides practical rules for manufacturing.

First, remember AQL is not a target defect rate for production—it’s the maximum defective percentage the receiving party is willing to accept for a sampled lot. AQL selection should reflect defect severity: critical defects usually carry an AQL of 0 (zero defects allowed), majors might use 0.65 to 1.5 for higher-risk items, and minors often sit at 2.5 or higher for low-consequence cosmetic issues. The standard sampling tables (ANSI/ASQ Z1.4 / ISO 2859) convert AQL and lot size to sample size and acceptance/rejection numbers.

Steps To Select AQL And Sample Size

1) Categorize Defects: Define critical, major, minor with objective criteria. 2) Assess Risk: For each category determine the business and safety impact of a missed defect. 3) Choose AQLs: Assign tighter AQLs to higher-impact categories. 4) Select Inspection Level: Choose I, II, III (II is general-purpose). 5) Use The Standard Table: Combine lot size, inspection level, and AQL to get sample size and acceptance numbers.

Common AQL Ranges And When To Use Them

• Critical Defects: AQL 0 or 0.01—use zero tolerance and consider 100% inspection for safety-related features.
• Major Defects: 0.65–1.5—used for functional failures that affect product performance.
• Minor Defects: 2.5–4.0+—used for cosmetic issues that do not affect function.

Inspection Level Selection

Inspection levels adjust sample sizes: Level I yields smaller samples, II is standard, and III is more stringent. Special levels (S-1 to S-4) reduce sample sizes for very small or low-risk lots. Choose level based on confidence in supplier quality and consequences of defects. For new suppliers or frequent failures, use level III to increase sample size and detection power.

How Lot Size Affects Sample Size

Standard tables scale sample sizes with lot quantity. Small lots may require minimal samples while very large lots plateau to larger but finite sample sizes—for example, many tables cap sample sizes around a few hundred units even for tens of thousands in a lot. Understand these limits when interpreting acceptance numbers for very large production runs.

Who Should Decide And How To Document

Decisions about AQL should be made jointly by the buyer’s quality team and the supplier during contract negotiation. Document AQLs, inspection level, defect definitions, sampling standard, and remedies for failure in the quality agreement or purchase order. Clear documentation prevents disputes and speeds resolutions when inspections fail.

Practical Example

A consumer electronics manufacturer orders 8,000 USB chargers. They specify: AQL 0 (critical for electrical safety), AQL 0.65 (major—functional), AQL 2.5 (minor—finish), inspection level II, ANSI/ASQ Z1.4. The sampling table returns a sample size of 200. For the critical class an AQL of 0 means any critical defect triggers rejection and further investigation.

Tips To Reduce Risk And Optimize Sampling

• Label: Start with conservative AQLs for new suppliers; relax AQLs as supplier performance improves and is documented.
• Label: Use inspection data to implement supplier corrective actions and process improvements rather than relying solely on tightened sampling.
• Label: Combine AQL sampling with SPC—sampling tells you lot acceptability, SPC finds process drift and root causes.
• Label: For costly defects, require 100% or multiple-stage inspections for critical features even if overall AQL is used for other attributes.

In short, the AQL Inspection level and sample size choices should match defect severity, supplier maturity, and the business consequences of failures; using standards, clear contracts, and data-driven adjustments will keep inspection efficient and effective.