The response surface methodology revisited – comparison of analytical and non-parametric approaches

The response surface methodology revisited – comparison of analytical and non-parametric approaches

Since G.E.P. Box introduced central composite designs in early fifties of 20th century, the classic design of experiments (DoE) utilizes response surface models (RSM), however usually limited to the simple form of low-degree polynomials. In the case of small size datasets, the conformity with the no...

Full description

Bibliographic Details
Main Authors: Przemysław Osocha, Jordan Podgórski
Format: Article
Language:English
Published: Printing House The Managers of Quality and Production Association 2018-09-01
Series:Production Engineering Archives
Subjects:
expert system
design of experiment
factorials
Taguchi robust design
RSM
Online Access:https://content.sciendo.com/view/journals/pea/20/20/article-p49.xml
id doaj-1d5f78af97654518b1ff2c5d737fa5d9
record_format Article
spelling doaj-1d5f78af97654518b1ff2c5d737fa5d92020-11-24T21:58:41ZengPrinting House The Managers of Quality and Production AssociationProduction Engineering Archives2353-51562353-77792018-09-0120(2018)495210.30657/pea.2018.20.10The response surface methodology revisited – comparison of analytical and non-parametric approachesPrzemysław Osocha0Jordan Podgórski 1Cracow University of TechnologyCracow University of TechnologySince G.E.P. Box introduced central composite designs in early fifties of 20th century, the classic design of experiments (DoE) utilizes response surface models (RSM), however usually limited to the simple form of low-degree polynomials. In the case of small size datasets, the conformity with the normal distribution has very weak reliability and it leads to very uncertain assessment of a parameter statistical significance. The bootstrap approach appears to be better solution than – theoretically proved but only asymptotically equal – t distribution based evaluation. The authors presents the comparison of the RSM model evaluated by a classic method and bootstrap approach. https://content.sciendo.com/view/journals/pea/20/20/article-p49.xmlexpert systemdesign of experimentfactorialsTaguchi robust designRSM
collection DOAJ
language English
format Article
sources DOAJ
author Przemysław Osocha
Jordan Podgórski
spellingShingle Przemysław Osocha
Jordan Podgórski
The response surface methodology revisited – comparison of analytical and non-parametric approaches
Production Engineering Archives
expert system
design of experiment
factorials
Taguchi robust design
RSM
author_facet Przemysław Osocha
Jordan Podgórski
author_sort Przemysław Osocha
title The response surface methodology revisited – comparison of analytical and non-parametric approaches
title_short The response surface methodology revisited – comparison of analytical and non-parametric approaches
title_full The response surface methodology revisited – comparison of analytical and non-parametric approaches
title_fullStr The response surface methodology revisited – comparison of analytical and non-parametric approaches
title_full_unstemmed The response surface methodology revisited – comparison of analytical and non-parametric approaches
title_sort response surface methodology revisited – comparison of analytical and non-parametric approaches
publisher Printing House The Managers of Quality and Production Association
series Production Engineering Archives
issn 2353-5156
2353-7779
publishDate 2018-09-01
description Since G.E.P. Box introduced central composite designs in early fifties of 20th century, the classic design of experiments (DoE) utilizes response surface models (RSM), however usually limited to the simple form of low-degree polynomials. In the case of small size datasets, the conformity with the normal distribution has very weak reliability and it leads to very uncertain assessment of a parameter statistical significance. The bootstrap approach appears to be better solution than – theoretically proved but only asymptotically equal – t distribution based evaluation. The authors presents the comparison of the RSM model evaluated by a classic method and bootstrap approach.
topic expert system
design of experiment
factorials
Taguchi robust design
RSM
url https://content.sciendo.com/view/journals/pea/20/20/article-p49.xml
work_keys_str_mv AT przemysławosocha theresponsesurfacemethodologyrevisitedcomparisonofanalyticalandnonparametricapproaches
AT jordanpodgorski theresponsesurfacemethodologyrevisitedcomparisonofanalyticalandnonparametricapproaches
AT przemysławosocha responsesurfacemethodologyrevisitedcomparisonofanalyticalandnonparametricapproaches
AT jordanpodgorski responsesurfacemethodologyrevisitedcomparisonofanalyticalandnonparametricapproaches
_version_ 1725850665820880896

Similar Items