摘要:

Variable Sampling Interval (VSI) control charts have been studied extensively, and it has been shown for Shewhart, cumulative sum (CUSUM), and exponentially weighted moving average (EWMA) charts that shifts in the process mean can be detected more quickly with no increase in the average in-control sampling or false alarm frequency relative to fixed interval charts. Applications of VSI charts in industry have been few, however, primarily due to logistical problems associated with a variable sample schedule. In this paper, we present an application of a VSI-EWMA chart at Monsanto's nylon fiber plant in Pensacola, Florida. Motivation for this implementation, instead of being quicker detection of shifts, was a significant decrease in laboratory cost with little adverse impact on control chart performance compared with the fixed-interval EWMA chart. The logistical problems were solved by employing a fixed time option with the VSI scheme (VSI-FT), where there is a fixed sample schedule for the long interval, and in-between samples are taken when warranted by the data.

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979607

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

摘要:

Advances in automated sampling technology have made autocorrelated data commonplace. Positive autocorrelation degrades control charts designed by classical methods. If a correct time-series model of the autocorrelated process is available, many have advocated the use of control charts on the residuals from the model. Using the average run length criterion in an AR(1) model, we show that plotting averages of batches of the raw data can be an effective alternative to plotting residuals. We consider both weighted averages and the simple, model-free approach of arithmetic averages. We compare these statistics to residuals in both Shewhart and cumulative sum (CUSUM) control charts.

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979608

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

摘要:

The sensitivity of four tests on ShewhartQcharts and of specially designed EWMA and CUSUMQcharts to detect one-step permanent shifts of a Poisson parameter has been studied. The usual test that signals for one point beyond three standard deviations on a Shewhart chart is found to have poor sensitivity, generally. The test that signals when four out of five consecutive points are beyond one standard deviation in the same direction on a ShewhartQchart is found to be a good omnibus test. The EWMA and CUSUMQcharts are most sensitive to detect this particular type of instability and are about comparable in overall performance.

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979609

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

摘要:

The geometric distribution based on the inverse binomial method of sampling is used to defineQstatistics for the case when the process falloutpis known and when it is not known in advance of sampling. The effectiveness to detect shifts inpof four tests made on ShewhartQcharts of these statistics as well as both exponentially weighted moving average and cumulative sumQcharts computed from these statistics is studied. The classic test of one point outside 3-sigma control limits is found to have poor sensitivity to detect a one-step permanent shift inp.The test consisting of four-out-of-five points beyond one standard deviation is found to be a good general omnibus test on a ShewhartQchart to detect an increase or decrease inp.

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979610

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979611

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979612

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979613

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979614

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979615

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor

摘要:

When the number of defectives (nonconforming products) is to be used as the response in two-level factorial and fractional factorial experiments, it is helpful to have an estimate of the sample size necessary for detecting a specific change. In this article we provide an easy-to-use table for standard 8, 16 and 32 run two-level factorial experiments that serves as a guideline for the necessary sample sizes to detect different changes in the average quality level. Examples of the use of the table are also provided.

ISSN:0022-4065    

DOI:10.1080/00224065.1995.11979616

出版商:Taylor&Francis    

年代:1995

数据来源: Taylor