摘要:

Evolutionary operation, as originally presented by G. E. P. Box in 1957, is now an accepted means of improving the performance of industrial processes. Numerous articles have since been published relating to the successful application of this procedure and they indicate that the technique is one of general industrial importance. It is the purpose of this article to review briefly these applications, thus providing a source of references useful both to those familiar with EVOP and those wishing to examine the potential applicability of the method for an existing process.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490372

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

This paper considers a model for the replenishment of communications satellites which may fail at random when in orbit. It is assumed that when the satellite level reaches a valueN, launchings are made until the level reachesM, at which time launchings are terminated. We calculate the probability thatnor fewer satellites are functional under this regime.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490373

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

This paper unifies and extends previously published characterizations of moving average, geometric moving average, and cumulative sum control chart procedures. It presents comparable characterizations of two procedures based on tests described but not evaluated in earlier papers. One of these procedures is based on a test devised by Girshick and Rubin that is optimal under a particular set of idealized conditions. The other procedure is based on run sum tests, which are generalizations of the type of run test that counts the number of consecutive points that exceed a limit, the generalization taking into account the extent that points in such a run exceed the limit.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490374

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

nobservations are taken from a mixture ofKnormal subpopulations, where the value ofKis known. It is assumed that thesenobservations are given asNfrequencies from equally spaced intervals. Initial guesses of theKmeans,Kvariances, andK− 1 proportions are made using the maximum likelihood estimates for a single truncated normal population as derived by Hald. Then an approximation to the likelihood function of the entire sample is used, and attempts to maximize this yield two iteration formulas. In practice, the method of steepest descent always converged, although the rate was not always fast. Special cases of equal variances and variances proportional to the square of the mean are also considered.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490375

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490376

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

The extreme vertices design is developed as a procedure for conducting experiments with mixtures when several factors have constraints placed on them. The constraints so imposed reduce the size of the factor space which would result had the factor levels been restricted to only 0 to 100 per cent. The selection of the vertices and the various centroids of the resulting hyper-polyhedron as the design is a method of determining a unique set of treatment combinations. This selection is motivated by the desire to explore the extremes as well as the center of the factor space. A non-linear programming procedure is used in determining the optimum treatment combination.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490377

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490378

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

Balanced, nested (hierarchal) experimental designs are widely publicized and used to isolate and estimate variances associated with multi-stage processes. Beyond the two-stage design, there has been no information on the distributions of the estimates of variance components. This paper presents some initial results, mostly empirical, dealing with the distributions of the estimates of variance components derived from 5 × 2 × 2 × 2 balanced nested design. The results of a sampling study using eight different combinations of input variances is described and discussed. Normal, rectangular, and exponential parent populations were studied separately. Uncertainty associated with variance component estimation is shown to be intolerably great under certain commonly occurring conditions. The frequency of occurrence of negative estimates of variance components is emphasized.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490379

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

After a 2n−kfractional factorial has been analysed, an experimenter may wish to carry out a further fraction to improve some estimates and break chains of aliases. The problem of augmenting a 2n−1half replicate by a nonoverlapping 2n−kdesign is discussed. In each alias set for the 2n−kfraction the effects occur in pairs as aliases from the half replicate. If only one of these pairs in a given alias set contains two nonnegligible effects, then that chain is broken. If more than one pair has both effects nonnegligible, no chains are broken. Procedures are given for augmenting four-letter plans for four or six factors.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490380

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor

摘要:

A method for construction of cumulative sum control charts for controlling the mean of a Weibull distribution is described. As a special case, charts appropriate to exponentially distributed variables can be constructed. There is some investigation of the results of using such charts when a non-exponential Weibull distribution would be more appropriate. The paper concludes with a discussion of certain formulas in the related analysis of sequential probability ratio tests for Weibull distributions.

ISSN:0040-1706    

DOI:10.1080/00401706.1966.10490381

出版商:Taylor & Francis Group    

年代:1966

数据来源: Taylor