Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
Multiattribute acceptance sampling plans - Library(ISI Kolkata ...
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
2.1 General cost models<br />
2.1.1 Scope<br />
So far we have discussed the comparative features of A kind, C kind and D kind multiattribute<br />
<strong>acceptance</strong> <strong>sampling</strong> schemes based on OC function. We now attempt to do this on<br />
the basis of economic considerations. We first review the various cost models for dealing simultaneously<br />
with more than one attribute. We further develop a generalized cost model as<br />
appropriate to compare the costs of different <strong>sampling</strong> schemes proposed by us, under the assumptions<br />
of independence of defect occurrences and also under the assumption of mutually<br />
exclusive occurrences of defect in a lot/sample. We do this under Poisson conditions.<br />
2.1.2 Existing cost models for multiattribute <strong>acceptance</strong> <strong>sampling</strong><br />
Hald [ See Chapter 0.2.7 ] has been the major contributor in the field of economic design<br />
of <strong>acceptance</strong> <strong>sampling</strong> <strong>plans</strong>. He obtained general solutions for linear cost models under<br />
discrete and continuous prior distribution of process average for single attribute. Hald’s<br />
major emphasis has been on finding asymptotic relationships between n and c, and between<br />
N and n for a single attribute single <strong>sampling</strong> plan.<br />
The economic design of multiattribute <strong>sampling</strong> scheme was considered by Schimdt and<br />
Bennett(1972), and further by Case, Schimdt and Bennett(1975), Ailor, Schimdt and Bennet(1975),<br />
Majumdar(1980), Moskowitz, Plante, Tang and Ravindran(1984), Tang, Plante<br />
and Mokowitz(1986), and Majumdar(1997). We discuss some of them and examine their<br />
relevance in the context of our enquiries.<br />
i) Case, Schimdt and Bennett(1975)<br />
This model is developed under following assumptions:<br />
(a) Each attribute is assumed to have its own sample size n i and an <strong>acceptance</strong> number c i<br />
for i = 1, 2, ..., r; r being the number of attributes.<br />
(b) Any item inspected on one attribute may be inspected on all other attributes thus resulting<br />
in the total member of items sampled being maximum of (n 1 , n 2 , ..., n r ). Acceptance<br />
is realized only if and only if x i ≤ c i ; i = 1, 2, ..., r<br />
(c) The number of items inspected for the i th attribute is without exception n i . No screening/sorting<br />
is made on the rejected lots.<br />
(d) Irrespective of the lot size a rejected lot is ‘scrapped’ at a fixed cost.<br />
(e) The sampled items are replaced in the lot by additional items and are taken from a lot<br />
of the same overall quality as the sampled lots.<br />
78