Sampling Acceptance Calculator

Use the fields below to evaluate a sampling acceptance plan, estimate risks, and compare expected outgoing quality metrics.

Example Data Table

This sample table shows how a quality team might compare different lots using one acceptance plan framework.

Formula Used

Acceptance decision: Accept the lot when observed defects d ≤ c. If d > c, reject or screen the lot.

Observed defect rate: p̂ = d / n, where d is observed defects and n is the sample size.

Acceptance probability (binomial model): P(Accept) = Σ [ C(n, x) p^x (1-p)^(n-x) ] for x = 0 to c.

Producer risk (α): α = 1 - P(Accept | AQL). This estimates the chance of rejecting a good lot.

Consumer risk (β): β = P(Accept | LTPD). This estimates the chance of accepting a poor lot.

AOQ: AOQ = p × P(Accept) × (N - n) / N under rectifying inspection assumptions.

ATI: ATI = n × P(Accept) + N × (1 - P(Accept)), giving the expected units inspected.

AOQL: The calculator scans incoming defect rates and returns the maximum AOQ value with the corresponding incoming defect rate.

How to Use This Calculator

1. Enter the lot size, sample size, and acceptance number from your inspection plan.
2. Type the observed number of defects found in the sample.
3. Provide AQL and LTPD percentages for your quality agreement or internal policy.
4. Select the probability model. Use hypergeometric when lot size effects matter.
5. Choose the inspection mode label and confidence level for the defect-rate interval.
6. Click Calculate Sampling Acceptance to display results above the form.
7. Use the CSV or PDF buttons to export the result summary.

Sampling Plan Design Benchmarks

A practical sampling acceptance plan starts with a clear risk target and lot profile. For stable production, teams often set AQL near 0.65% to 1.5% and LTPD near 4% to 8%, depending on customer criticality. Sample size selection should balance inspection cost and risk separation. Larger samples improve discrimination between good and poor lots, but increase handling time, labor demand, and inspection fatigue across shifts. Trend data strengthens plan revisions.

Interpreting Producer and Consumer Risk

Producer risk alpha measures how often a good lot is rejected at the AQL level. Consumer risk beta measures how often a poor lot is accepted at the LTPD level. In many industrial programs, alpha is targeted near 5% and beta near 10%, though regulated sectors may require tighter limits. Reviewing both values prevents decisions based only on observed defects without probability context. This keeps supplier penalties aligned with actual process capability.

Using Acceptance Probability for Operations

Acceptance probability at the observed defect rate helps supervisors decide whether current performance is likely to pass future lots under the same plan. If the observed acceptance probability is high while beta remains elevated, the plan may be too permissive for customer protection. If acceptance probability collapses after minor defect increases, the plan may be too strict for process variation and throughput stability. Use trend charts to compare consecutive lots and shifts.

AOQ AOQL and ATI Decision Value

AOQ estimates outgoing defect rate under rectifying inspection assumptions, while AOQL shows the worst expected outgoing quality across incoming defect levels. These metrics are useful for budgeting rework and screening capacity. ATI estimates expected units inspected, which directly affects staffing and cycle time. Comparing ATI between plans can reveal lower total inspection effort even when the sample size is larger initially. Pair ATI estimates with labor rates to forecast inspection budget impact.

Implementation Controls and Review Cadence

Teams should standardize how defect classes are counted before using any calculator output. Repeated disagreements on defect coding will distort observed rates, confidence intervals, and risk estimates. A monthly review of accepted and rejected lots, false rejects, and customer returns helps recalibrate AQL, LTPD, and sample size. Document model choice as well, because hypergeometric assumptions matter when sampling fractions are large. Controlled reviews improve consistency, traceability, and supplier communication during external audits.

FAQs

1) What is an acceptance number?

It is the maximum defects allowed in the sample for lot acceptance. If observed defects exceed this value, the lot is rejected or routed for screening based on your procedure.

2) Which model should I choose?

Use hypergeometric when the sample is a meaningful share of the lot. Use binomial for large lots with small sampling fractions. Poisson works as a quick approximation at low defect rates.

3) How are AQL and LTPD different?

AQL represents acceptable incoming quality for producer-risk evaluation. LTPD represents poor incoming quality for consumer-risk evaluation. The calculator uses both to estimate how the plan behaves at two control points.

4) Why can a lot be accepted with defects found?

Single sampling plans allow some defects up to the acceptance number. The decision reflects statistical risk and inspection economics, not a claim that the lot is defect-free.

5) What does AOQL help me manage?

AOQL estimates the worst expected outgoing defect level under rectifying inspection assumptions. It helps teams compare sampling plans, rework burden, and outgoing quality protection before changing inspection rules.

6) Does this replace formal standards tables?

No. It supports planning, training, and scenario analysis. Use your approved standard, customer contract, and internal quality procedures when assigning code letters, switching rules, or compliance documentation.