4.10 Determination of Acceptability.
4.10.1 Percent Defective Inspection.
To determine acceptablity of a lot or batch under percent defective
inspection, the applicable sampling plan shall be used in accordance with 4.10.1.1, 4.10.1.2, 4.10.1.3, and
4.10.1.4.
4.10.1.1 Single Sampling Plan.
The number of sample units inspected shall be equal to the sample size given by
the plan. if the number of defectives found in the sample is equal to or less than the acceptance number, the lot
or batch shall be considered acceptable. If the number of defectives is equal to or greater than the rejection
number, the lot or batch shall be rejected.
4.10.1.2 Double Sampling Plan.
Anumber of sample units equal to the first sample size given by the plan shall be
inspected. If the number of defectives found in the first sample is equal to or less than the first acceptance
number, the lot or batch shall considered acceptable. If the number of defectives found in the first sample is
equal to or reater than the first rejection number, the lot or batch shall be rejected. If the number of
defectives found in the first sample is between the first acceptance and rejection numbers, a second sample of
the same size shall be inspected. the number of defectives found in the first and second samples shall be
accumulated. If the cumulative number of defectives is equal to or less thean the second acceptance number,the
lot or batch shall be considered acceptable. If the cumulative number of successive samples required to reach a
decision my be as seven.
4.10.1.3 Multiple Sample Plan.
Under multiple sampling, the procedure shall be similar to that specified in 4.10.1.2,except that the number of successive samples required to reach a decision may be as many as seven.
4.10.1.4 Special Procedure for Reduced Inspection.
Under reduced inspection, the sampling procedure may terminate
without either acceptance or rejection criteria having been met. In these circumstances, the lot or batch will be
considered acceptable, but normal inspection will be reinstated startin with the next lot or batch (see 4.7.4.b).
4.10.2 Defects per Handred Units Inspection.
To determine the acceptability of a lot or btch under defects per
hundred units inspection, the procedure specified for percent defective isnpection above shall be used, except
that the word "defects" shall be substituted for "defectives.
4.11 Limiting Quality Protection.
The sampling plans and associated procedures given ins this publication were
designed for use where the units of product are produced in a continuing series of lots or batches over a period
of time. However, if the lot or batch is of an isolated nature, it is desirable to limit the selection of
sampling plans to those, associated with a designated AQL value, that provide not less than a specified limiting
quality protection. Sampling plans for the purpose can be selected by choosing a Limiting Quality (LQ) and a
consumer's risk to be associated with it. Tables VI and VII give values of LQ for the commonly used consumer's
risks of 10 percent and 5 percent respectively. If a different value of consumer's risk is required, the O.C.
curves and their tabulated value of consumer's risk is required, the O.C. curves and their tabulated values may
bu used. The concept of LQ may also be useful in specifying the AQL and Inspection Levels for a series of lots or
batches, thus fixing minimum sample size where there is some reason for avoiding (with more than a given
consumer's risk) more than a limiting proportion of defectives (or defects) in any single lot or batch.
4.12 Curves
4.12.1 Operating Characteristic Curves.
The operating characteristic curves for normal inspection, shown in Table
X,indicate the percentage of lots or batches which may be expected to be accepted under the various sampling
plans for a given process quality. The curves shown are for single sampling; curves for double and multiple
sampling are matched as closely as practicable. The O.C. curves shown for AQLs greater than 10.0 are based on the
Poisson distribution and are applicable for defects per hundred units inspection; those for AQLs of 10.0 or less
and sample sizes of 80 or less are based on the binomial distribution and are applicable for percent defective
inspection; Those for QQLs of 10.0 or less and sample sizes larger than 80 are based the Poisson distribution and
are applicable either for defects per hundred units inspection, or for percent defective inspection (the Poissson
distribution being an adequate approximation to the binomial distribution under these conditions). Tabulated
values, corresponding to selected values or probabilities of acceptance (Pa, in percent) are given for each of
the curves shown, and, in addition, for tightened inspection, and for defects per hundred units for AQLs of 10.0
or less and sample sizes of 80 or less.
4.12.2 Average Sample Size Curves.
Average sample size curves for double and multiple sampling are Table IX.
These show the average sample sizes which may be expected to occur under the various sampling plans for given
levels of process quality. The curves assume no curtailment of inspection and are approximate to the extent that
they are based upon the Poisson distribution, and that the sample sizes for double and multiple samplin are
assumed to be 0.631n and 0.25n respectively, where n is the equivalent sample size.
. |
