Statistical Engineering Examples in the Engine Oil Additive Industry

ABSTRACT

Over the past several years, through papers, presentations, and articles, the concept of statistical engineering has been presented, defined, and described. However, there appear to be very few answers to the Hoerl and Snee question/challenge (posed at the 2010 Joint Statistical Meetings) as to who is practicing statistical engineering and building something out of current statistical methods.

According to Hoerl and Snee (2010), statistical engineering is the study of how to best utilize statistical concepts, methods, and tools and integrate them with information technology and other relevant sciences to generate improved results. They also aptly described it as trying to build something with the statistical science “parts list” of methods.

Statistical engineering should not be equated with applied statistics. Both applied and theoretical statistics are parts of statistical engineering, but in order to be considered statistical engineering, the parts must create a whole system and/or protocol that is more than the sum of the parts. In addition, a high-level need of an organization must be met, and the process must have a sustainable life.

Lean Six Sigma (LSS) is cited as an example of statistical engineering. It involves and uses statistical aspects such as experimental design, variance component analysis, and control charting; meets the need for an organization to improve quality; and is a continuous practice, day to day and year to year, without any special reinvention. However, LSS is not the only example of this concept.

The engine oil industry in North America operates through several different committees, organizations, and trade associations. Over the past 20 years, statisticians from ally and rival companies have participated in numerous subgroups and panels to create procedures, systems, practices, and protocols that serve the greater good of the industry as well as the end consumer. Statistical methods, tools, and logic form the basis for these practices and protocols that were put into place to solve problems and meet the needs of the engine oil industry and member organization engineers, chemists, and marketers. These systems are an important part of the process of how engine oil is tested, approved, and ultimately put into a motor vehicle. These systems are also excellent examples of statistical engineering in practice.

KEYWORDS:

• statistical engineering
• statistical practice
• statistical tools

ACKNOWLEDGMENTS

I thank all of the professionals who have contributed to the development and enhancement of statistical engineering in the North American engine oil industry. I would especially like to thank the statisticians in our industry for their dedication and integrity.