Introduction to special edition of Quality Engineering

This special issue of Quality Engineering is dedicated to the topic of statistical engineering (SE). The primary motivation for the development of SE as a discipline has been the gap between the impact statistics has had on society, which has been noteworthy to be sure, and the impact that it could and should have, which would be transformational. Ironically, ten years ago Christine Anderson-Cook and Lu Lu saw this issue, and co-edited the first special edition of Quality Engineering on SE (Anderson-Cook and Lu 2012), which at that time was just beginning to emerge as a unique discipline. A lot has happened between 2012 and 2022, hence we feel that this is an excellent time to revisit SE in the journal that has published many of the seminal papers on the topic. In their introduction to the first special edition, Anderson-Cook and Lu stated:

We, the editors of this volume and the authors, believe that SE has an emerging and substantial role to play in the evolution of our profession. Statisticians have long been developing tools and methods in a diverse array of areas, but there has been little formal structure guiding how these can be combined to solve more complicated and important problems. Some applied statisticians have taught themselves how to select, sequence, and synthesize tools to benefit their organizations and impact the bottom line, but there has been little formally available to leverage this expertise and disseminate it to the broader community.

Ten years later, it is reasonable to ask how far the profession has come in “disseminating expertise” in how to “select, sequence, and synthesize tools” to address complex and important problems. How has the underlying theory of how to accomplish this synthetization progressed since 2012? Further, what case studies have been produced to illustrate how this can be accomplished in practice? Is there evidence of tangible impact from SE? We anticipate that this special edition will help answer these critical questions, as least to some degree.

Before discussing the individual articles, we should provide a brief introduction to SE for those who may not be familiar with this discipline. First of all, in the phrase “statistical engineering,” “engineering” is the noun, that is, the “what.” “Statistical” is the adjective, which modifies that noun, i.e., explains the type of engineering. In other words, SE is a form of engineering, one in which statistics plays a heavy role. Of course, nouns and adjectives cannot be reversed without changing the meaning of the phrase. Therefore, just as “data science” is not the same thing as “scientific data,” “engineering statistics” is not the same thing as SE. Engineering statistics refers to the application of statistics to engineering problems, while SE refers to the engineering of solutions to complex statistical problems, which might be in healthcare, finance, education, or any other application area.

While the term “statistical engineering” goes back to Eisenhart (1950), the two of us began using the term to refer to engineering solutions to complex statistical problems in 2010 (Hoerl and Snee 2010). As noted previously, we felt that statistical methods could and should have much broader impact on society. The International Statistical Engineering Association (ISEA, isea-change.org) subsequently provided the current definition commonly in use: “The discipline of statistical engineering is the study of the systematic integration of statistical concepts, methods, and tools, often with other relevant disciplines, to solve important problems sustainably.” There are obviously some important terms in this definition that warrant elaboration:

• SE is a discipline, i.e., the study of something, not a set of tools.

• As an engineering discipline SE does not focus on advancing the fundamental knowledge of the physical world, i.e., it is not a science. Rather, as with other engineering disciplines, it utilizes existing concepts, methods and tools in novel ways to achieve novel results. In this sense it is complementary to statistical science, just as chemical engineering is complementary to chemistry.

• Concepts, methods and tools are each important and need to be integrated. This integration should be done in a systematic rather than ad-hoc manner.

• Other relevant disciplines, such as computer science, are often required to solve complex statistical problems. Subject-matter knowledge is typically key to success.

• As an engineering discipline, the ultimate goal of SE is to solve important problems. While this may seem obvious, an emphasis on solving important problems provides SE with perhaps its most important attribute, being tool-agnostic. Having loyalty to individual tools distracts problem-solvers from selecting the most appropriate tools to address a given problem.

• Further, SE seeks solutions that are sustainable. We argue that many solutions, including those published in professional journals, provide technical solutions. But all too frequently, these solutions are not sustainable over time, typically because problem-solvers did not consider the full complexity of the problem, including organizational and political complexity (Hoerl, Jensen, and de Mast 2021).

For those readers of this special edition motivated to study statistical engineering further, we of course recommend the 2012 special edition mentioned previously. We also recommend the Statistical Engineering Handbook (Hare et al. 2021), which is published by ISEA, and is available on the members-only section of the ISEA website (isea-change.org). Note that membership in ISEA is free to individuals, hence the handbook can be obtained at no charge. While there are numerous other publications on the topic worth reading, we would in particular recommend:

• Hoerl and Snee (2017), which provides a detailed explanation of how SE is unique as a discipline, and

• Hoerl and Vining (2021), which details the historical development of SE, including the founding of ISEA as a professional society solely dedicated to SE.

The articles in this special edition are diverse, and cover important territory related to the theory and practice of SE. Christine Anderson-Cook and Lu Lu, who as we noted previously co-edited the first special edition on SE, provide a panel discussion for this special edition on the past, present, and future of SE. They organized a very impressive group of panelists, representing some of the original developers of SE, as well as some professionals who are fairly new to SE. They obtained so much material that the panel discussion is split into a discussion of the past and present, and another focused on the future of SE.

Susan Schall provides a comparison of SE with more established engineering disciplines, and notes how synergistic they can be. As a trained and experienced industrial engineer who has an extensive background in statistical engineering, she is perhaps in a unique position to provide this perspective.

Tom Redman, Diego Kuonen, and Roger provide an elaboration of the project framing phase of SE projects. They explain why poor project framing is often a root cause of project failure, and how this phase should be properly addressed.

Per our previous discussion of providing case studies to illustrate how SE plays out in practice, we have four case studies, each conducted in a different environment. Lynne Hare provides a case showing how SE principles can be deployed throughout an organization. This study was conducted primarily in Mexico, demonstrating the universal applicability of SE. Ron provides a case study from pharmaceutical development, noting the need to address the full range of challenges from technical to legal to organizational. A team from P&G provides another international case study involving development and standardization of consumer projects on a global basis. Lastly, a team from JMP provides a case study illustrating the application of SE to software development.

We trust that you will find this diverse array of articles on SE as interesting and informative as we do. Further, we anticipate that you will see how proper application of statistical engineering, in conjunction with statistical science and other disciplines, can help our profession drive broader impact for society. Best wishes,

Roger W. Hoerl
Union College
Ronald D. Snee
Snee Associates