Inspection by Attributes

Statistical Sampling Plan, OC Curve & Average Outgoing Quality

A sampling plan does not measure quality. It sorts lots, and the OC curve says how well.

The Lot Size and inspection severity
units

Level II unless the contract says otherwise

The Plan Acceptance criterion and the process it will meet
%
defects

Take from ISO 2859-1 Table 2-A for contractual work

%

Sample Size

— units

ISO 2859-1 code letter for this lot size and level

Operating Characteristic & Inspection Load

Acceptance Probability at the AQL
— %
Producer's Risk
— %
Quality Rejected 9 Times in 10 (LTPD)
— %
Indifference Quality (50% Accept)
— %
Average Outgoing Quality
— %
Average Outgoing Quality Limit
— %
Average Total Inspection
— units
Rejection Number Re
— defects

The sample size comes from the ISO 2859-1 general inspection level code letters, with Level I and Level III taken as one code letter either side of Level II - which is how the general levels are laid out in Table 1. Special levels S-1 to S-4, used for destructive or expensive tests, are not covered and must be taken from the standard. The acceptance number is an input, not a lookup: Table 2-A contains arrow cells that redirect to a different sample size, and reproducing those without the table in front of you is how plans get misquoted. The operating characteristic is exact binomial for a single sampling plan under normal inspection; double, multiple and sequential plans have different curves. Where the sample would exceed the lot it is capped at the lot size, which is 100% inspection. Switching rules between normal, tightened and reduced inspection are a separate mechanism and change the effective protection substantially.

Statistical Sampling Plan, OC Curve & Average Outgoing Quality — free, with the formula and a worked example, at Textile School.