• Communications for Statistical Applications and Methods
• /
• v.8 no.2
• /
• pp.383-394
• /
• 2001

This paper is a generalization of single and two-level skip-lot sampling plans to n-level. On every skipping inspection of the n-level skip-lot sampling plan, not only the number of consecutive lots to be accepted but also the fraction of lots to be inspected can be freely choosed. The general formulas of the operating characteristic function, average fraction inspected, average sample number and average outgoing quality in n-level skip-lot sampling plan are derived. The operating characteristic curves, average sample number and average outgoing quality of a reference plan, two-level and five-level skip-lot sampling plans are compared.

• Communications for Statistical Applications and Methods
• /
• v.3 no.2
• /
• pp.151-159
• /
• 1996

The general formulas of average fraction inspected, average sample number and average outgoing quality in n-level skip-lot sampling plan are derived. Average sample number and average outgoing quality of a reference plan, three-level, five-level and ten-level skip-lot sampling plans are compared.

• IE interfaces
• /
• v.24 no.1
• /
• pp.87-96
• /
• 2011

In this paper, we address an optimization problem for minimizing the inspection and rework cost in an inspection-rework system, which forms a network of nodes including a K-stage inspection system, storage areas for items, a source inspection shop, and a re-inspection shop. We assume that (n, 0) acceptance sampling is performed in the source inspection shop and that only 100(1-r)% of items of rejected lots are re-inspected in the re-inspection shop. Since all the nodes are interrelated, in order to formulate our steady-state objective function, we make a steady-state network flow analysis between nodes, and derive both the steady-state amount of flows between nodes and the steady-state fraction defectives by solving a nonlinear balance equation. Finally we provide some fundamental properties and an enumeration procedure for determining the optimal values of (K, r) which both minimizes our objective function and attains a given target average outgoing quality.

• Industrial Engineering and Management Systems
• /
• v.4 no.1
• /
• pp.75-81
• /
• 2005

Borror et al. discussed the EWMA(Exponentially Weighted Moving Average) chart to monitor the count of defects which follows the Poisson distribution, referred to the $EWMA_c$ chart, as an alternative Shewhart c chart. In the $EWMA_c$ chart, the Markov chain approach is used to calculate the ARL (Average Run Length). On the other hand, in order to monitor the process fraction defectives P in high-yield processes, Xie et al. presented the CCC(Cumulative Count of Conforming)-r chart of which quality characteristic is the cumulative count of conforming item inspected until observing $r({\geq}2)$ nonconforming items. Furthermore, Ohta and Kusukawa presented the $CS(Confirmation Sample)_{CCC-r}$ chart as an alternative of the CCC-r chart. As a more superior chart in high-yield processes, in this paper we present an $EWMA_{CCC-r}$ chart to detect more sensitively small or moderate shifts in P than the $CS_{CCC-r}$ chart. The proposed $EWMA_{CCC-r}$ chart can be constructed by applying the designing method of the $EWMA_C$ chart to the CCC-r chart. ANOS(Average Number of Observations to Signal) of the proposed chart is compared with that of the $CS_{CCC-r}$ chart through computer simulation. It is demonstrated from numerical examples that the performance of proposed chart is more superior to the $CS_{CCC-r}$ chart.