Acceptance Sampling and Risk
Acceptance Sampling and Risk
Calculates acceptance probabilities of a user-defined sampling plan under the binomial model.
How to use the result
Random, independent inspections with a constant probability of a defective item are assumed. For a small finite lot sampled without replacement, a hypergeometric calculation is required.
Defective items are counted, not multiple defects on one item. The inspection itself is assumed to be error-free.
n and c are entered freely. No normative AQL tables, lot-acceptance decisions, switching rules, or ISO-2859 sampling plans are generated.
Formulas
Pa = Σ from j=0 to c: n! / (j! × (n−j)!) × p^j × (1−p)^(n−j)α = P(X > c | p_good); β = P(X ≤ c | p_bad)E(X) = n × p; Divide percentage values by 100 before calculation. Sources: https://www.itl.nist.gov/div898/handbook/pmc/section2/pmc232.htm