No abstract

No abstract

In this paper, we present two new dependent mixed sampling plans and a mixed repetitive sampling plan based on the process yield index for linear profiles. A mixed sampling plan includes two stages; the first stage is a combined test by variables and by attributes, and the second stage is tested by attributes only. The first plan is based on the traditional mixed sampling scheme without marginal quality, and the second plan is based on the modified mixed sampling scheme with marginal quality. If a lot is very bad, the probability it can be rejected with the small initial sample is higher for dependent mixed sampling with marginal quality. We compare dependent mixed plans with other types of double sampling plans; the former outperforms the others with respect to average sample number. Additionally, the number of profiles required in the first stage is much smaller than in the second stage. When the inspection is costly and destructive, a mixed repetitive sampling plan includes one stage that combines the inspection by attributes and by variables and allows resampling in some conditions. We compare the mixed repetitive sampling plan with the repetitive sampling plans by attributes and by variables. The probability of acceptance using the mixed repetitive sampling plan performs better than the repetitive sampling plans by attributes and by variables. Three examples are used to illustrate the proposed methods. Copyright © 2017 John Wiley & Sons, Ltd.

In many industrial applications, the quality of a process or product can be characterized by a function or profile. Owing to spatial autocorrelation or time collapse, the assumption of the observations within each profile that are uncorrelated is violated. This paper aims at evaluating the process yield for linear within-profile autocorrelation. We present an approximate lower confidence bound for S pkA when the observations within each profile follow a first-order autoregressive AR(1) model. A simulation study is conducted to assess the performance of the proposed method. The simulation results confirm that the proposed method performs well for the bias, the standard deviation, and the coverage rate. One real example is used to demonstrate the applications of the proposed approach.Note: bias, estimated value-true value; std, standard deviation; average CI, average confidence interval of yield; and LCB is obtained at the 95% confidence level.

The quality of a process is characterized by a functional relationship between the response variable and one or more explanatory variables called a linear profile. In this study, the process selection problem that deals with comparing two linear profiles with one‐sided specifications and selecting the one that has a significantly higher capability value is presented. An exact approach based on the ratio test statistic to tackle the supplier selection problem is proposed. Testing hypotheses with two cases for comparing two processes is considered. Critical values of the tests are calculated to determine the selection decisions. The number of profiles required for a designated selection power and confidence level is also provided. The results provide useful information to practitioners. One real example is used to illustrate the application of our proposed approach. Copyright © 2014 John Wiley & Sons, Ltd.