Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
0.4 Scope of the present inquiry<br />
0.4.1: The multiattribute <strong>sampling</strong> inspection<br />
At the outset, we take into account the fact that the number of elements of the set of characteristics<br />
to be verified, for the purpose of <strong>acceptance</strong>/rejection of lots or batches of mass<br />
produced items, is unlikely to be just one in most of the practical situations. For example, a<br />
sample of finished garment is verified for 15 characteristics (all attribute type) before delivery.<br />
Hansen (1957) has listed 130 attributes for automotive final inspection and road test.<br />
We attend to the problems specific to formulating lot-by-lot multiattribute <strong>sampling</strong> inspection<br />
schemes in such situations.<br />
0.4.2 The existing method of choosing <strong>sampling</strong> inspection <strong>plans</strong> based on seriousness<br />
of characteristics.<br />
The quality characteristics are decidedly unequal in their effect on fitness for use. A relatively<br />
few are serious i.e. of critical importance; many are minor. Clearly, the more important the<br />
characteristic, the greater the attention it should receive in such matters as: extent of quality<br />
planning, precision of processes, tooling and instrumentation, sizes of samples, strictness of<br />
conformance etc. To this end many companies utilize formal systems of seriousness classification.<br />
Juran and Gryna(1996) tabulate one such classification scheme in food industry.<br />
The general practice of the industry has been to assign different AQL values and employ<br />
effectively a parallel system of <strong>sampling</strong> inspections. For example, Hansen (1957) reported<br />
adaptation of MIL-STD-105A to the <strong>acceptance</strong> inspection of the M38A1 truck commonly<br />
called ‘the jeep’ manufactured by Wiley’s Motors Inc, in Toledo, Ohio. Two hundred and<br />
four characteristics were classified in four classes, namely: special defects ( hundred per cent<br />
inspection ) comprising of 11 attributes, major defects ( AQL 15 per one hundred vehicle<br />
) comprising of 14 attributes, minor defects (AQL 150 defects per 100 vehicle) comprising<br />
of 69 attributes, incidental defects ( 400 defects per 100 vehicles) comprising of 110 attributes.<br />
The MIL-STD-105D recommends classification of defects and designate different AQLs<br />
for groups of defects or for individual defects as the case may be. However, since sample size<br />
is taken as a function of lot size (for most of the time), the <strong>acceptance</strong> number get affected<br />
by variable AQL’s. For lot sizes around 50, for the vehicle example, we take a sample of size<br />
8 ( inspection level II, normal inspection) and inspect for all three classes of attributes and<br />
accept the ‘jeep’, if the number of major defects is less than or equal to 3, the number of<br />
minor defects is less than 15 and the number of incidental defects is less than 44.<br />
10