Analysis considerations in industrial split-plot experiments with non-normal responses

Timothy J. Robinson, Raymond H. Myers, Douglas Montgomery

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Non-normal responses are common in many industrial experiments. When there are factors whose levels are difficult and/or costly to control, an experiment is typically run within a split-plot context. Industrial split-plot experiments have received a great deal of attention in the literature in the normal response, linear model setting. Generalized linear models have been proposed for the analysis of completely randomized designs when the response is non-normal. When one uses a completely randomized design, there is an implication that responses are independent. For split-plot experimentation there is an assumption that responses within a given whole plot are correlated. To account for this correlation, generalized linear mixed models (GLMM) can be used for analysis. Generalized linear mixed models fall into two categories: population-averaged models and batch-specific models. We give an expository treatment of GLMM and discuss the differences in interpretations of the population-averaged and batch-specific models using an example from film manufacturing.

Original languageEnglish (US)
Pages (from-to)180-192
Number of pages13
JournalJournal of Quality Technology
Volume36
Issue number2
StatePublished - Apr 2004

Fingerprint

Split-plot
Generalized Linear Mixed Model
Experiment
Experiments
Batch
Generalized Linear Model
Population Model
Experimentation
Linear Model
Manufacturing
Model
Generalized linear mixed model

Keywords

  • Generalized Linear Mixed Models
  • Mixture Experiments
  • Split-plot Designs

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Statistics and Probability
  • Management Science and Operations Research

Cite this

Analysis considerations in industrial split-plot experiments with non-normal responses. / Robinson, Timothy J.; Myers, Raymond H.; Montgomery, Douglas.

In: Journal of Quality Technology, Vol. 36, No. 2, 04.2004, p. 180-192.

Research output: Contribution to journalArticle

@article{28e2df6cb297453db7bad42d3fadb287,
title = "Analysis considerations in industrial split-plot experiments with non-normal responses",
abstract = "Non-normal responses are common in many industrial experiments. When there are factors whose levels are difficult and/or costly to control, an experiment is typically run within a split-plot context. Industrial split-plot experiments have received a great deal of attention in the literature in the normal response, linear model setting. Generalized linear models have been proposed for the analysis of completely randomized designs when the response is non-normal. When one uses a completely randomized design, there is an implication that responses are independent. For split-plot experimentation there is an assumption that responses within a given whole plot are correlated. To account for this correlation, generalized linear mixed models (GLMM) can be used for analysis. Generalized linear mixed models fall into two categories: population-averaged models and batch-specific models. We give an expository treatment of GLMM and discuss the differences in interpretations of the population-averaged and batch-specific models using an example from film manufacturing.",
keywords = "Generalized Linear Mixed Models, Mixture Experiments, Split-plot Designs",
author = "Robinson, {Timothy J.} and Myers, {Raymond H.} and Douglas Montgomery",
year = "2004",
month = "4",
language = "English (US)",
volume = "36",
pages = "180--192",
journal = "Journal of Quality Technology",
issn = "0022-4065",
publisher = "American Society for Quality",
number = "2",

}

TY - JOUR

T1 - Analysis considerations in industrial split-plot experiments with non-normal responses

AU - Robinson, Timothy J.

AU - Myers, Raymond H.

AU - Montgomery, Douglas

PY - 2004/4

Y1 - 2004/4

N2 - Non-normal responses are common in many industrial experiments. When there are factors whose levels are difficult and/or costly to control, an experiment is typically run within a split-plot context. Industrial split-plot experiments have received a great deal of attention in the literature in the normal response, linear model setting. Generalized linear models have been proposed for the analysis of completely randomized designs when the response is non-normal. When one uses a completely randomized design, there is an implication that responses are independent. For split-plot experimentation there is an assumption that responses within a given whole plot are correlated. To account for this correlation, generalized linear mixed models (GLMM) can be used for analysis. Generalized linear mixed models fall into two categories: population-averaged models and batch-specific models. We give an expository treatment of GLMM and discuss the differences in interpretations of the population-averaged and batch-specific models using an example from film manufacturing.

AB - Non-normal responses are common in many industrial experiments. When there are factors whose levels are difficult and/or costly to control, an experiment is typically run within a split-plot context. Industrial split-plot experiments have received a great deal of attention in the literature in the normal response, linear model setting. Generalized linear models have been proposed for the analysis of completely randomized designs when the response is non-normal. When one uses a completely randomized design, there is an implication that responses are independent. For split-plot experimentation there is an assumption that responses within a given whole plot are correlated. To account for this correlation, generalized linear mixed models (GLMM) can be used for analysis. Generalized linear mixed models fall into two categories: population-averaged models and batch-specific models. We give an expository treatment of GLMM and discuss the differences in interpretations of the population-averaged and batch-specific models using an example from film manufacturing.

KW - Generalized Linear Mixed Models

KW - Mixture Experiments

KW - Split-plot Designs

UR - http://www.scopus.com/inward/record.url?scp=2342594557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2342594557&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:2342594557

VL - 36

SP - 180

EP - 192

JO - Journal of Quality Technology

JF - Journal of Quality Technology

SN - 0022-4065

IS - 2

ER -