摘要:

Manufacturing processes seldom run perfectly. In attempts at improvement, these processes are continually“adjusted”by the mechanic, supervisor, and engineer; or more formally, they are“studied”from manufacturing, quality control, or engineering perspectives. These studies may be generically categorized as process quality optimization studies. Four such studies are defined, explained, and compared in this article. These are the process performance check, the process performance evaluation, the process capability study, and the process improvement program. In addition it is shown how these studies are used, separately or in combination, to achieve process quality optimization systematically, effectively, and at minimum cost.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980949

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

Addressed is the problem of determining a one-sided tolerance limit for a population possessing a distribution belonging to a broad class of lifetime distributions. A new implementation of existing general theory is given and contrasted with an earlier utilization of that theory. General guidelines are given for deciding which implementation to use. A method for adjusting for the accuracy of the measuring device is discussed and illustrated with an example.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980950

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

A nomogram based on the method of moments is given for estimating the three parameters of the Weibull distribution. Iterative procedures for parameter estimation using digital computers require initial estimates of the parameters. An empirical procedure useful in providing these estimates is described. Simulated data obtained from Weibull populations provides examples illustrating the use of both the nomogram and the empirical procedure.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980951

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

An economic analysis is made of skip-lot sampling plans for both destructive and non-destructive testing. The purpose of this analysis is to find an optimal single sampling plan, to be characterized bynandcand skip-lot parametersiandf, which minimizes the average cost per unit. The procedure for finding an optimal inspection plan by computer search facilitates the numerical evaluation.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980952

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

This paper displays and describes a table useful in a variety of quality control estimation problems where the underlying distribution is binomial. Sample size is, in every case, less than or equal to 32; censoring is permitted in the data. Two examples of the use of the table are given.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980953

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

An apparent paradox occurring in statistical hypothesis testing is described and resolved. What seem to be contradictions are shown to be no more than the result of confusion on the part of the experimenter as to what conclusion should be drawn from an arbitrarily stated null hypothesis. A confidence interval approach that avoids this confusion is also described.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980954

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

The United States Pharmacopeiapresents sampling plans and specifications for pharmaceutical products which are used in judging the quality of these products. Previously accepted and revised sampling plans for product dissolution are evaluated and compared using conventional statistical techniques and computer simulation. This paper presents operating characteristic curves and related graphs for these dissolution sampling plans that form a basis upon which lots of dosage units are accepted or rejected. These plans have also been evaluated for their effectiveness in assuring the quality of drug products.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980955

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

A computer program for performing two diagnostic procedures suggested by Daniel and Wood for a multiple regression model is presented. The first procedure is designed to detect points that are remote inx-space and that may be overly influential on the model summary statistics. The second procedure allows the analyst to identify pairs of points that are near neighbors inx-space and to compute an estimate of pure error using the residuals at these points. Two examples are given.

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980956

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980957

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1980.11980958

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor