Non‐linear nonparametric quality screening in low sampling testing

Non‐linear nonparametric quality screening in low sampling testing Purpose – Screening simultaneously for effects and their curvature may be useful in industrial environments when an economic restriction on experimentation is imposed. Saturated‐unreplicated fractional factorial designs have been a regular outlet for scheduling screening investigations under such circumstances. The purpose of this paper is to devise a practical test that may simultaneously quantify in statistical terms the possible existence of active factors in concert with an associated non‐linearity during screening. Design/methodology/approach – The three‐level, nine‐run orthogonal design is utilized to compute a family of parameter‐free reference cumulative distributions by permuting ranked observations via a brute‐force method. The proposed technique is simple, practical and non‐graphical. It is based on Kruskal‐Wallis test and involves a sum of effects through the squared rank‐sum inference statistic. This statistic is appropriately extended for fractional factorial composite contrasting while avoiding explicitly the effect sparsity assumption. Findings – The method is shown to be worthy competing with mainstream comparison methods and aids in averting potential complications arising from the indiscriminant use of analysis of variance in very low sampling schemes where subjective variance pooling is otherwise enforced. Research limitations/implications – The true distributions obtained in this paper are suitable for sieving a fairly small amount of potential control factors while maintaining the non‐linearity question in the search. Practical implications – The method is objective and is further elucidated by reworking two recent case studies which account for a total of five saturated screenings. Originality/value – The statistical tables produced are easy to use and uphold the need for estimating separately mean and variance effects which are rather difficult to pinpoint for the fast track, low‐volume trials this paper is intended to. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Quality & Reliability Management Emerald Publishing

Non‐linear nonparametric quality screening in low sampling testing

Loading next page...
 
/lp/emerald-publishing/non-linear-nonparametric-quality-screening-in-low-sampling-testing-hXifM60EPV
Publisher
Emerald Publishing
Copyright
Copyright © 2010 Emerald Group Publishing Limited. All rights reserved.
ISSN
0265-671X
D.O.I.
10.1108/02656711011075107
Publisher site
See Article on Publisher Site

Abstract

Purpose – Screening simultaneously for effects and their curvature may be useful in industrial environments when an economic restriction on experimentation is imposed. Saturated‐unreplicated fractional factorial designs have been a regular outlet for scheduling screening investigations under such circumstances. The purpose of this paper is to devise a practical test that may simultaneously quantify in statistical terms the possible existence of active factors in concert with an associated non‐linearity during screening. Design/methodology/approach – The three‐level, nine‐run orthogonal design is utilized to compute a family of parameter‐free reference cumulative distributions by permuting ranked observations via a brute‐force method. The proposed technique is simple, practical and non‐graphical. It is based on Kruskal‐Wallis test and involves a sum of effects through the squared rank‐sum inference statistic. This statistic is appropriately extended for fractional factorial composite contrasting while avoiding explicitly the effect sparsity assumption. Findings – The method is shown to be worthy competing with mainstream comparison methods and aids in averting potential complications arising from the indiscriminant use of analysis of variance in very low sampling schemes where subjective variance pooling is otherwise enforced. Research limitations/implications – The true distributions obtained in this paper are suitable for sieving a fairly small amount of potential control factors while maintaining the non‐linearity question in the search. Practical implications – The method is objective and is further elucidated by reworking two recent case studies which account for a total of five saturated screenings. Originality/value – The statistical tables produced are easy to use and uphold the need for estimating separately mean and variance effects which are rather difficult to pinpoint for the fast track, low‐volume trials this paper is intended to.

Journal

International Journal of Quality & Reliability ManagementEmerald Publishing

Published: Sep 7, 2010

Keywords: Experimental design; Quality; Statistical testing

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off