TRY THIS TODAY

# Don’t Lose The Recipe

## Identify and document knacks to keep processes running smoothly

by Matthew Barsalou and Robert Perkin

Robert’s Grandma Perkin was famous in her family for her butterscotch pie. It was a staple of Thanksgiving dinner for as long as anyone could remember. But no one ever thought about exactly how she made it or what would happen if she stopped making it.

Eventually, Grandma Perkin passed away, and so did her beloved recipe. Robert’s sister slaved to recreate the recipe, but to no avail.

This family story illustrates a problem that, unfortunately, is all too common in a manufacturing environment. For some reason, a process that has functioned just fine for years starts failing and no one can figure out why. Suddenly, the recipe is lost.

Often, this happens when a long-time production employee leaves the organization and takes his or her special recipe, or method, which almost always is undocumented.

Such an employee’s superior way of doing something is called a knack. An employee who knows from experience to file down a critical component on a system that is being assembled, for example, has a knack. Other employees don’t know the part must be filed because the action was never documented, resulting in higher scrap for employees without the knack.1

To solve this problem, the knack must be identified and documented. But first, the root cause analysis investigator must discover that there is a difference between employees that potentially could result from a knack.

A good first step in an investigation is to graph the data and look for salient features.2 Salient features stand out and may be related to the cause, but they warrant further investigation because often they provide clues during a root cause analysis. Online Figure 1 shows a run chart of a manufacturing process’ total scrap. There are good days and bad days, but nothing stands out.

Stratification can provide additional insight into the data3 and should be used when the data consist of an aggregate from multiple sources. The run chart in Online Figure 1 consists of the total scrap of four employees. The run chart in Online Figure 2 shows the same data stratified by employee. Stratification shows that Charles has a significantly lower scrap rate than the others. So, what is Charles doing differently?

To help identify the knack, it may be necessary to observe all four employees and create a process flowchart of every move they make during the assembly operation. After the knack is found, the procedure should be revised.

Documentation must be specific and actionable at the operator level, otherwise the knowledge may be lost when the employee with the knack no longer is around to provide guidance. After the knack is found and documented, formally document the "recipe" to ensure consistent success.

### References

1. Frank M. Gryna, Quality Planning and Analysis: From Product Development Through Use," fourth edition, McGraw-Hill, 2001.
2. Jeroen de Mast and Albert Trip, "Quality-Improvement Projects," Journal of Quality Technology, Vol. 39, No.4, 2007, pp. 301-311.
3. Kaoru Ishikawa, Guide to Quality Control, second edition, Asian Productivity Organization, 1991.

Matthew Barsalou is a statistical problem resolution Master Black Belt at BorgWarner Turbo Systems Engineering GmbH in Kirchheimbolanden, Germany. He has a master’s degree in business administration and engineering from Wilhelm Büchner Hochschule in Darmstadt, Germany, and a master’s degree in liberal studies from Fort Hays State University in Hays, KS. Barsalou is an ASQ senior member and holds several certifications.

Robert Perkin is the senior manager for engineering problem resolution and statistical methods for BorgWarner Turbo Systems. Perkin has master’s degrees in management of technology and engineering management from Washington University in St. Louis. He is a Smarter Solutions-certified lean Six Sigma Master Black Belt.

This is really good article. I will use it to train my staff in finding out the knack and sharing the pie!
--Tanmay, 11-18-2017

Out of 1 Ratings