Sequential and systematic sampling plans for the Markov-dependent production process

Sequential and systematic sampling plans for the Markov-dependent production process

0.00 Avg rating0 Votes
Article ID: iaor20021678
Country: United States
Volume: 48
Issue: 6
Start Page Number: 451
End Page Number: 467
Publication Date: Sep 2001
Journal: Naval Research Logistics
Authors: ,
Keywords: markov processes
Abstract:

Acceptance sampling plans are used to assess the quality of an ongoing production processs, in addition to the lot acceptance. In this paper, we consider sampling inspection plans for monitoring the Markov-dependent production process. We construct sequential plans that satisfy the usual probability requirements at acceptable quality level and rejectable quality level and, in addition, possess the minimum average sample number under semicurtailed inspection. As these plans result in large sample sizes, especially when the serial correlation is high, we suggest new plans called ‘systematic sampling plans’. The minimum average sample number systematic plans that satisfy the probability requirements are constructed. Our algorithm uses some simple recurrence relations to compute the required acceptance probabilities. The optimal systematic plans require much smaller sample sizes and acceptance numbers, compared to the sequential plans. However, they need larger production runs to make a decision. Tables for choosing appropriate sequential and systematic plans are provided. The problem of selecting the best systematic sampling plan is also addressed. The operating characteristic curves of some of the sequential and the systematic plans are compared, and are observed to be almost identical.

Reviews

Required fields are marked *. Your email address will not be published.