摘要:

Successful analysis of superlarge data sets requires statistical procedures that automatically clean the data and uncover simple structure. The protocol described applies to multivariate industrial data from continuous manufacturing processes with feedback and feedforward control. Our methods form a twelve-step sequence that edits and re-lags the time series, as well as applying diagnostics to look for subtle data flaws. At different stages, the protocol will reject data, impute data, re-lag the time series, flag categories of suspicious data, and divide the data set into more homogeneous subsets. The result is a clean data set ready for analysis using standard statistical packages or software tools. Although there is no guarantee that every corruption has been caught and corrected, the output data set is more thoroughly examined than traditional human-intensive methods can achieve. To assist in this preliminary analysis, four graphical methods are described, which were developed in studies of glass manufacture from PPG Industries' production plants and sheet aluminum production by Alcoa.

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979382

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

Traditionally, Plackett-Burman (PB) designs have been used in screening experiments for identifying important main effects. The PB designs whose run sizes are not a power of two have been criticized for their complex aliasing patterns, which according to conventional wisdom gives confusing results. This paper goes beyond the traditional approach by proposing an analysis strategy that entertains interactions in addition to main effects. Based on the precepts of effect sparsity and effect heredity, the proposed procedure exploits the designs' complex aliasing patterns, thereby turning their“liability”into an advantage. Demonstration of the procedure on three real experiments shows the potential for extracting important information available in the data that has, until now, been missed. Some limitations are discussed, and extensions to overcome them are given. The proposed procedure also applies to more general mixed level designs that have become increasingly popular.

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979383

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

When the usual assumption of constant variances does not hold, the recommended substitute for a deviations from nominal (DNOM) chart is to chart standardized deviations. This has the added difficulty of estimating different standard deviations for each subgroup. As an alternative to this, an approach to DNOM charts that allows more general process and measurement error models is presented. When applied to the constant variance case, the usual chart of differences from nominal results. When applied to the constant coefficient of variation case, a new DNOM chart based on ratios of subgroup means to nominals results. Both charts have the desirable feature of constant control limits.

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979384

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

Truncated data arise when a variable is observable only over some portion of its range. This paper describes how truncated data arise in studies of the field performance or reliability of manufactured items. Failure to account for truncation can lead to biased inferences. Some useful nonparametric methods are presented with examples.

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979385

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

A FORTRAN program is presented that computes new performance measures for continuous sampling plans applied to finite production runs. For specified clearance number and sampling fraction, the program calculates the maximum average outgoing quality and maximum percentiles for the outgoing quality (under both rectifying and nonrectifying inspection). It also calculates means, variances, and percentiles for the outgoing quality, for the fraction inspected, and for an estimator of incoming quality. The latter results can be used to calculate confidence intervals for the incoming quality and prediction and tolerance intervals for the outgoing quality using inspection data. These new performance measures are intended to supplement the tables in MIL-STD-1235B for use in selecting CSP-1 plans.

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979386

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979387

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979388

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979389

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979390

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1992.11979391

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor