Six Sigma (6σ) is a process improvement methodology that has been adopted by leading companies throughout the world. I’ll not review it here, because a simple Google search will return about 8 million hits, most of them for companies more than willing to sell you a $10,000 training course to help certify you as a “black belt” in just a few weeks. Most college students struggle to learn even basic statistics in a 16-week semester, but somehow you can become an expert in all aspects of statistics, quality control, design of experiments, and process improvement in a three-week workshop. It makes me wonder why I spent all that time getting my engineering degrees.

Proponents of 6σ claim it was “invented” by Motorola. Yet, all of the 6σ books I have on my shelf seem to be filled with the work of pioneers like Sir Ronald Fisher, Walter Shewhart, W. Edwards Deming, Joe Juran, Taiichi Ohno, Shigeo Shingo and George Box. Their work has been widely published for years in many books, and taught in a variety of statistics and industrial engineering curricula. I’m not sure why the knowledge they created had to be aggregated and called “Six Sigma.” I guess they were pretty smart, but apparently just didn’t know how to package their stuff.

Proponents of 6σ will also tell you that you can reach a state of nirvana when you improve your process to a point where it produces only 3.4 defective parts per million (ppm). With a little digging, you will find that this magical number is the area under a normal curve outside of a point that is 4.5 standard deviations away from the mean. The complete foolishness of this calculation is monumental, and I could go on a long rant about it, but that is probably best left as a topic for another article.

So, what’s wrong with striving for a really low defect rate like 3.4 ppm? After all, defects like damaged products, late shipments and mispicks are costly! Of course, but my problem is that 3.4 ppm is an arbitrary line in the sand that won’t apply to most processes.

Let’s examine what would happen if we used 3.4 ppm as a goal for the airline industry. Between 1970 and 2006 there were 146,970,000 airline flights in the U.S. and Canada, and 58 experienced an “incident.” Some of these were minor, like sliding off an icy runway with no one hurt, while others were major crashes in which people were killed. That works out to a rate of only 0.395 incidents per million flights. Had the airline industry been better trained in 6σ, they would have stopped improving their systems long ago, and we would have experienced nearly 10 times more incidents.

Now, don’t get me wrong, there are a lot of good techniques and great people employed in formal 6σ programs. Topics such as focusing on customer needs, process measurement, leadership, employee involvement, data-driven decision making, organization-wide quality education, continuous improvement and most of all, variance reduction are great things to do. But you don’t need a formal 6σ program to apply these methods to your materials handling processes. I encourage you to learn about them and hire people who have been well trained in their application, even if their belts aren’t certifiably black.

So what does the future hold for Six Sigma? Based on the patterns of the last 25 years or so, I expect we will soon see some sort of repackaging—a re-branding effort if you will. So far we’ve had TQM and BPR, (three-letter acronyms), ISO-9000 (a three letter acronym with a number) and now, 6σ (a number and a Greek letter). Hmmm…maybe the next big thing will be just a special symbol of some sort, like the singer Prince tried. You won’t be able to refer to it by name, but somehow you’ll just know it’s the right thing to do.