Too Soon to Sigma?


By Robert Minardi

While attending Vision Expo West in 2016, I had dinner with some industry regulars that have more than a century of combined experience in the optical field. The topic of Six Sigma briefly came up, and I mentioned that it doesn’t seem to be a priority in the optical industry.

One of my dinner companions said, “The really big labs have guys that do that, but the smaller labs don’t really need it.”

I was a little taken aback by this reply at first. However, then I asked myself: “Does a lab need to be a certain size to implement Six Sigma?”

Before we get into that, we should define what Six Sigma is. There are two definitions. One is statistical and the other is the methodology.

What is a “Six Sigma?”

The statistical representation of a Sigma is determined by the standard deviation formula:

Without getting “mathy,” a Sigma level describes how close you are to perfection. The higher the Sigma level, the closer you are to having zero defects, as illustrated in the following table:

The Sigma Level on the left corresponds to average number of defects you would generally see in one million production opportunities—which is not a completed job but rather an opportunity (or production step) during which a defect may occur; in fact, one process step may have many defect opportunities. For example, surface blocking defects may include off-axis, unwanted prism, eye-4-eye and scratches; among others. That’s at least four opportunities for a defect at one workstation.

The resulting yield, then, is the percentage of “good products” produced.

To put sigma levels in perspective, if you played two rounds of golf per week, a two sigma level of play would mean missing about six putts per round. Not bad. According to Nick Wreden, in the book Profit Brand: How to Increase the Profitability, Accountability and Sustainability of Brands, a Six Sigma level of play would mean you missed a putt every 163 years! Tiger who?

Basically, a Six Sigma level of defects is darn near perfect. Admittedly, this probably isn’t attainable for an optical lab. So, what are we talking about here?

The Method

The Six Sigma methodology was introduced in the mid-1980s by Motorola engineer Bill Smith. The goal of was to diagnose and eliminate variation in a process, thereby reducing defects and allowing you to provide the highest-quality, lowest-cost product to your customer.

Jack Welch of GE defined the methodology as follows: “Six Sigma is a quality program that, when all is said and done, improves your customer’s experience, lowers your costs, and builds better leaders.” Welch implemented the program at GE in late 1995, and by 1999 was named “Manager of the Century” by Fortune magazine. During his tenure, GE’s value rose a staggering 4,000 percent.

The DMAIC Approach

DMAIC (pronounced dah-MAY-ik) is an acronym for the five phases of a Six Sigma project: Define, Measure, Analyze, Improve and Control. These phases are carried out, one after another, until the project’s conclusion. The number of tools available for each phase is vast. Let’s just look at a very basic definition of each phase.

The Define Phase: During the define phase, you determine what your issues are based on the voice of the customer. Their complaints and concerns should be the driving force behind project selection.

The following story demonstrates why this is important: Company XYZ makes spot-welded metal shelves. The engineers at XYZ noticed bumps on the sides of the shelves caused by the welding process. In an effort to get rid of these bumps, the company spent millions of dollars in time and materials with little progress. One day, a customer arrived and the lead engineer points out they’ve started to see progress on the weld bump issue. The customer looked at the engineer and said “Weld bumps? I don’t care about the bumps. I came here to talk about the front of the shelves being crooked!”

This is a humorous, albeit cringe-worthy, example of the importance of listening to the customer’s needs.

Interestingly, a customer doesn’t have to be the actual customer. Anywhere materials are handed from one person to another is usually considered a supplier/customer relationship. For example, in the bench department, final inspection is the customer of insertion. If the inspectors are complaining about lenses being smudged from insertion, you could consider that a customer complaint.

The Measure Phase: The objectives during the measure are to collect data and measure what you can do right now versus where you need to be. For example, if you need to move jobs through a given area faster, you need to find out how long it currently takes and how long it should take to meet customer demands. Detailed and accurate data is essential here.

The Analyze Phase: At this stage, you analyze the data collected to identify where your issues are. Let the data tell you the root cause of the problem. During this phase, it’s important to separate the “vital few from the trivial many.” Focus on the fewest improvements that will reap the biggest reward.

The Improve Phase: The improve phase is all about developing a solution. This is where you use the knowledge gained from the previous steps. You know where the variation’s coming from, now it’s time to put your heads together and fix it.

The Control Phase: In the final phase, you implement the solution and set up a system to ensure it stays in place. 


In the eyecare profession, we use SOAP (Subjective, Objective Assessment and Plan) as a systematic approach for addressing patient issues. At its core, Six Sigma and DMAIC are the same thing; a step by step method that gives you a guided path to success.

So, when’s the best time to use these techniques? Now! It doesn’t matter if you produce 100 jobs a day or 3,000.

If you’re asking yourself “Why can’t I just collect some numbers and find the problem without all the other stuff?” What I’ve found is, a facility will be strong in some areas, but not others. For example, a lab may have a well-defined issue, but data collection isn’t aggressive, detailed or accurate enough to uncover the root cause. Another common situation arises when a problem is solved, but a solid control system isn’t implemented to make sure it doesn’t happen again.

This is where you end up with seasonal breakage spikes. There’s a reason the automotive, aerospace and technology industries use these techniques religiously. They work.

If you still think your lab isn’t “big” enough for Six Sigma, just remember, small labs grow into big labs. Sometimes quicker than you think.

Robert Minardi, ABOC, has been in manufacturing for almost 25 years. He’s a certified Lean Six Sigma Black Belt with a background in quality control.  

SIDEBAR: Here’s a couple of my favorite links to get you started:





Labtalk June 2020