On the Podium…
It's the Olympics, and I, like a lot of other people, am losing
a lot of sleep watching the television coverage from the
winter events in South Korea. I find myself particularly
attracted to the Alpine events and hockey. My university,
Arizona State University, has only recently fielded a sur-
prisingly competitive Division I NCAA men's hockey team,
and this likely has rekindled my interests in the sport.
The Olympic season has inspired me to examine the
recently completed 2017 Volume 33 of this journal to try
to identify my choices as the “medalists” from among the
numerous excellent papers that have appeared. Now a note
of warning. I have not used any objective judging criteria:
citations, subject matter, no collaboration with my co‐chief
editors, no advice from a panel of experts. Nothing but my
own probably highly biased opinion. I am not going to
award gold, silver, or bronze medal rankings either. I am
just going to list the papers in reverse order of publication
date (issue number), offer a few brief comments about each
paper, and let you, the reader, award the medal positions.
Or you can choose to completely disagree with my choices,
in which case please email me and offer your selections as
an alternative. They might appear in a future editorial.
The most recently published paper is by Janj and
Anderson‐Cook (2017). Their paper studies the effective-
ness of 2 popular techniques for analyzing the data from
unreplicated factorial designs, the normal (or equivalently
the half‐normal) probability plot, and a least absolute
shrinkage and selection operation plot. They investigate
how these methods perform when either 1 observation
is missing or when there is an outlier in the data. Because
these designs are so widely used in practice, the authors'
work should be of wide interest to industrial statisticians.
The next paper is Capaci, Bergquist, Kulahci, and
Vanhatalo (2017), which considers using designed experi-
engineering process control. There is a well‐established lite-
rature on the integration of statistical process monitoring
with engineering process control, but little work has been
reported on using designed experiments in these situations.
The authors study 2 situations. One explores the impact of
experimental factors that may be considered as disturbances
in a closed‐loop system. The second studies an experiment
using the set‐points of the controllers as the design factors.
Examples of how to analyze the 2 situations are presented.
The last paper is Steward and Rigdon (2017). Risk
adjustment is widely used in monitoring health care out-
comes. These procedures take into account measures of
the patient condition and how these measures are related
to the outcomes. When the outcome is dichotomous, such
as survival/death, the modeling involves logistic regres-
sion to assess the relationship between the predictor(s)
and the outcome. Most risk‐adjusted control charts are
designed to detect a change in the log‐odds of the adverse
outcome. These authors address risk‐adjusted monitoring
as a change‐point problem with several possible change‐
point models. Their approach generalizes previous risk‐
adjusted charts in that they try to detect changes in any
of the underlying model parameters. They adopt a Bayes-
ian approach and find the posterior distribution for the
model (ie, which coefficients changed), the time of the
change, and the values of the parameters for those that
changed. All 3 tasks are accomplished in the context of
a single model. They apply reversible jump MCMC to
account for the variable size of the parameter space.
All 3 of these papers are typical of the types of work
that this journal hopes to attract. They address important
practical challenges, they adopt modern methods to
address the problems, and they are clearly written so that
the work is accessible to the widest possible audiences.
Douglas C. Montgomery
Arizona State University,
Janj D‐H, Anderson‐Cook CM. Examining robustness of model
selection with half‐normal and LASSO plots for unreplicated fac-
torial designs. Quality and Reliability Engineering International.
Capaci F, Bergquist B, Kulahci M, Vanhatalo E. Exploring the use of
design of experiments in industrial processes operating
under closed‐loop control. Quality and Reliability Engineering
Steward RM, Rigdon SE. Risk‐adjusted monitoring of healthcare
quality: model selection and change‐point estimation.
Qual Reliab Engng Int. 2018;34:277. Copyright © 2018 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/qre 277