Compare rules
Fictional rules inspect n random units and accept at c = 0 defectives; otherwise reject. Assume a very large lot relative to n and binomial approximation. For stipulated defective fraction p, acceptance is (1 − p)^n; rejection is 100% minus acceptance. At 1%, 1 − p = 0.99.
n: 50; 100
Accept at 1%: 60.50%; 36.60%
Reject at 1%: 39.50%; 63.40%
Accept at 5%: 7.69%; 0.59%
Treat 1% as acceptable and 5% as unacceptable. Increasing n lowers customer risk of accepting 5% lots but raises producer risk of rejecting 1% lots. These quality inputs are not estimates from zero observed defects. Permissible risk limits remain unspecified; neither rule is endorsed.
Check the population
Define the homogeneous lot from which units will be selected. Minitab calls for random selection representing the entire lot. A convenient accessible subset does not establish that basis. Record whether the result counts defective units, counts multiple defects, or measures a characteristic: those data types require different treatment. For an isolated finite lot with pass/fail data, Minitab identifies the hypergeometric distribution as the exact acceptance calculation. A large-lot approximation is not automatically its exact answer. Ask the quality reviewer which population, data and model the proposal actually concerns before treating the computed numbers as applicable.
Check the procedure
A named standard can describe more than a standalone sample size. ASQ’s public overview describes Z1.4 as a system with switching rules for a continuing stream of lots, including normal, tightened and reduced inspection. It does not provide the operative tables here. If a proposal cites that standard, ask which applicable procedure and inspection state the quality specialist is using. A sample-size quotation alone does not establish that the system has been applied. This guide neither applies the paid standard nor substitutes its illustration for the required procedure.
Record R17
R17, fictional lot: compare both rules above. Quality points stipulated; agreed risk limits absent. Observed count unknown; representative sampling unconfirmed because rear cartons are inaccessible. Reviewer: quality lead. Obtain the selection basis, observed result, agreed risk limits and applicable procedure/state before this record supports disposition. No release authorized.
For general information and education, not legal, tax, investment or valuation advice. Examples are illustrative and do not predict your business’s value, financing terms or sale outcome. Consult qualified advisers about your situation.
Subscribe to Deal Flow Canada for practical owner guides and Canadian deal updates.