Evaluate lot acceptance using sample size, accept, and reject numbers. Determine pass/fail decisions for incoming and final inspections.
Acceptance sampling is a statistical quality control technique used to determine whether a production lot meets quality requirements without inspecting every unit. By drawing a random sample and counting defects, you compare the result against predetermined accept and reject numbers to make a lot disposition decision.
The approach balances inspection cost against the risk of accepting low-quality lots or rejecting good ones. A well-designed sampling plan specifies the sample size (n), the maximum number of defects to accept the lot (Ac), and the minimum number of defects to reject the lot (Re, typically Ac + 1).
This calculator lets you enter your sampling plan parameters and the actual defects found, then tells you whether to accept or reject the lot, along with the observed defect rate.
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.
Acceptance sampling provides an objective, data-driven lot disposition method that is faster and cheaper than 100% inspection while providing defined levels of quality assurance. Having accurate figures readily available streamlines reporting, audit preparation, and strategic planning discussions with management and key stakeholders across the business. Consistent measurement creates a reliable baseline for tracking improvements over time and demonstrating return on investment for process optimization initiatives.
Decision Rule: • If defects found ≤ Ac → Accept the lot • If defects found ≥ Re → Reject the lot Observed Defect Rate = Defects Found / Sample Size × 100%
Result: Accept the lot
With 2 defects found in a sample of 125, and Ac = 3: since 2 ≤ 3, the lot is accepted. The observed defect rate is 2/125 = 1.6%.
Single sampling plans are simplest: one sample, one decision. Double sampling allows a second sample if the first is inconclusive, reducing average sample size. Multiple and sequential sampling plans extend this concept further for maximum efficiency.
Producer's risk (α) is the probability of rejecting a lot at the AQL quality level — typically set at 5%. Consumer's risk (β) is the probability of accepting a lot at the LTPD quality level — typically set at 10%. The OC curve visualizes these risks.
Randomize sample selection within each lot. Document and standardize defect definitions. Use attribute sampling for pass/fail characteristics and variables sampling for measured dimensions. Review and update plans periodically as process quality improves.
Single sampling inspects one fixed sample and makes a decision. Double sampling first inspects a smaller sample; if results are borderline, a second sample is taken. Double sampling can reduce average inspection effort.
The lot is accepted. The accept number (Ac) is the maximum number of defects allowed for acceptance. Only when defects ≥ Re (typically Ac + 1) is the lot rejected.
No. Acceptance sampling is probabilistic. There is always a consumer's risk (β) that a bad lot is accepted and a producer's risk (α) that a good lot is rejected. These risks are controlled by the plan design.
Use 100% inspection when: defects pose safety risks, the cost of passing a defect far exceeds inspection cost, the lot is very small, or automated inspection is feasible and cost-effective. Running this calculation with a range of plausible inputs can help you understand the sensitivity of the result and plan for different scenarios.
Apply separate sampling plans for critical, major, and minor defects, each with different AQL levels. A lot must pass all three plans to be accepted.
Skip-lot sampling inspects only a fraction of submitted lots based on a supplier's demonstrated quality history. If a supplier passes many consecutive lots, some lots are "skipped" to reduce inspection cost.