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.
Table 1.2.4 Summary table, giving information about slope comparisons for the<br />
286 <strong>sampling</strong> <strong>plans</strong> with 3 classes of attributes<br />
Conditions<br />
Number of MASSP’s satisfying the given condition in the range:<br />
0 < p ≤TotalAQL 0 < p ≤ p (3) 0.1<br />
Number of <strong>plans</strong> with Slope 1 < 9 37<br />
Slope 2<br />
Number of <strong>plans</strong> with Slope 2 < 2 161<br />
Slope 3<br />
Number of <strong>plans</strong> with Slope 1 < 0 25<br />
Slope 2 and Slope 2 < Slope 3<br />
Number of <strong>plans</strong> with Slope 1 ><br />
Slope 2 > Slope 3<br />
275 113<br />
It is clear from the above table that, while using different AQL values as appropriate<br />
to the respective importance of the attributes and choosing the corresponding <strong>acceptance</strong><br />
numbers there is no guarantee that we would get a good plan in terms of risks as well as the<br />
discriminating power in a multiattribute situation. As regards the ratio p (3) 0.1/ TotalAQL,<br />
it has ranged from 1.7 to 12.<br />
We now try to generalize the above results more formally for a class of <strong>plans</strong> where<br />
a sample of given size n is chosen and the <strong>acceptance</strong> numbers are different for different<br />
attributes.<br />
1.2.5 The C kind plan and the properties of its OC Function<br />
We first note all these MASSP’s are based on the following <strong>acceptance</strong> criterion. We take<br />
a sample of size n, observe the number of defects for the i th attribute as x i and apply the<br />
following <strong>acceptance</strong> criterion, accept if: x i ≤ c i for all i = 1, 2, ..., r; reject otherwise.<br />
We will henceforth call these <strong>plans</strong> as C kind <strong>plans</strong>. The OC function under Poisson<br />
conditions is given by<br />
r∏<br />
P C(c 1 , c 2 , ..., c r ; m 1 , m 2 , ..., m r ) = G(c i , m i );<br />
i=1<br />
m i = np i ∀i<br />
47