Far East Journal of Theoretical Statistics
Volume 7, Issue 1, Pages 35 - 43
(May 2002)
|
|
A TWO DECISION CRITERIA FOR BAYESIAN MULTIPLE DEFERRED PLAN
WITH DISCRETE PRIOR DISTRIBUTION
K. K. Suresh (India) and O. S. Deepa (India)
|
Abstract: Risk based sampling plans are traditional in nature, drawing
upon producer and consumer types of risks as depicted by the operating
characteristic curve. When the between lot variation due to (non-removable)
assignable causes is large compared with within-lot variation, Bayesian
acceptance sampling should be considered for designing of any sampling plans.
Bayesian sampling plan takes into account the past history of similar lots
submitted previously for inspection. When producer and consumer are negotiating
quality limits and sampling plan, it is important especially for the producer to
find out the proportion of lots expected to be accepted under the plan when it
is in operation.
This paper presents a procedure for two decision criteria Bayesian Sampling
Plan based on Multiple Deferred State Sampling MDS(0, 1) Plan. The selection of
plan parameters for MDS(0, 1) plan through Bayes solution considering double
binomial as prior distribution is presented. An illustration is provided for an
easy selection of plan parameters. |
Keywords and phrases: multiple deferred state sampling, Bayes solution, double binomial prior
distribution, break-even quality level. |
|
Number of Downloads: 4 | Number of Views: 767 |
|