Top Six Sigma Tools to Use for Big Results: Part 1
As a Six Sigma practitioner and mentor, I am often asked what Six Sigma tool to use to solve specific quality problems. The answer is, "It depends." After all, at last count there were over 100 various tools available for the Six Sigma professional to use (see Nancy R. Teague’s The Quality Toolbox, second edition). I have learned over the years, however, that the simplest tools often are not only the most appropriate, but also provide the biggest returns with their use. 99 times out of 100, the Six Sigma tool of choice is one of the seven basic quality tools.
You as a Six Sigma practitioner should master each of these seven tools. Once you do, you will find yourself referring to them not as "basic" but as "indispensable."
Top Seven Six Sigma Tools
Here are the seven basic quality tools everyone should keep in his or her toolbox:
- Check sheet
- Cause and effect diagram
- Control chart
- Pareto chart
- Scatter diagram
I will explore the first four tools; the remaining three tools I will discuss in Part 2 of this article series.
I know what you’re thinking. "A check sheet? Come-on, what can you do with that?" Let me show you. First, a check sheet is a structured form for collecting data when the data can be observed and collected repeatedly by the same person or at the same location. Second, the power of a check sheet lies in its simplicity; anyone can use one with few to no instructions, and it is not intimidating (like, say, a linear regression analysis).
A great example of the power of a check sheet is at Johns Hopkins University. The University boiled down a 64-page federal document on controlling hospital-acquired infections into five simple steps. In an 18-month study published in the New England Journal of Medicine (December 2006), the Johns Hopkins check sheet saved an estimated 1,500 lives and nearly $200 million. Simple, powerful, effective.
"Wow," you say. "Tell me more!" Well, I’m glad you’re enthusiastic. There are powerful examples like the one for Johns Hopkins for each of these basic tools. But for now, I’ll briefly review cause and effect diagrams, control charts, and Pareto charts.
When using the Pareto chart with your quality teams, throw in this tidbit of history and you will look like an expert. (Whether you really are is for you to know and others to find out.) The chart is named after Vilfredo Pareto (yes, his name is really Vilfredo, not Alfredo), an Italian economist and philosopher who studied, among other things, income distribution. The Pareto Principle was named after him due to his observation that 80 percent of the land in Italy was owned by 20 percent of the population. "Ah ha, so that’s where the 80-20 rule came from!" — Yep, it’s from our buddy Vilfredo.
Now that you are schooled on the history, what about the chart? The Pareto chart is really a fancy bar chart. The length of each bar represents the frequency of whatever it is you are counting. The chart is arranged with the longest bars on the left and the shortest bars on the right. Using this method, it is easy to pinpoint the most significant items. On top of each bar is the percentage each bar contributes to the total. The percentages are added as you move to the right until they reach 100 percent. In the example shown here, Defect 1 represents 42 percent of the count, Defect 2 is 32 percent, Defect 3 is 16 percent and Defect 4 is 10 percent.
Use the Pareto chart when you need to analyze data about the frequency of a problem, when looking at many problems or causes (and you want to focus on the most significant one), or when you are trying to narrow broad items into specific components.
Cause and Effect Diagram
This is a very cool and very effective tool for finding the root cause of a problem. (It’s also called an Ishikawa diagram, named for its inventor — I’ll spare you the history — and a fishbone diagram, because when completed, it resembles the bones of a fish.) Unlike the check sheet, however, you will have to train folks on how to use it. It’s a trivial price to pay for such big gains.
The diagram reads right to left (Ishikawa was from Japan, and Japanese is read right to left). The effect, or problem, is on the far right. The causal factors are listed on the left.
My experience is that you will need to guide participants to the major categories of causes to the problem. If you run into trouble, use generic headings like machines, material, process, people, and environment.
Write these categories on the branches pointing to the main arrow. Facilitate brainstorming for all possible causes of the problem. Ask "Why did this cause happen?" and list the answers as branches. Continue this process until all the causal factors are identified. Weight each causal factor based on the impact to the overall problem and then report your findings to the management team, along with your recommendations, of course.
The control chart is by far one of my favorite tools. Control charts are used to track measurements over time. It is widely considered to be the most powerful tool to analyze variation in processes.
This type of charting was originated by Walter Shewhart in the 1920s. His publication of Economic Control of Quality of Manufactured Product made the charts popular.
In a control chart, outliers are easy to spot. The ability to determine if your process is in control is relatively simple. (A process is in control when all the plotted points fall within the bounds of the control limits, and the points do not display any non-random patterns.) When a process is in control, it is predictable.
With control charts almost anyone can spot a trend. (Although it may take some time to teach managers that two data points in a row is not a trend.) Here are a few rules to follow when looking at control charts:
- Anything outside the control limits is considered "special" and a cause needs to be identified.
- Seven or more consecutive points in a row on the same side of the centerline is considered a run violation and should be investigated.
- A trend is when five or more consecutive points move upward or downward, or seven or more points drift in a single direction. (You need twenty plot points or more before the seven rule is valid.)
Remember, big payoffs come from using simple tools. In the next installment of this series, I will cover the scatter diagram, stratification, and the histogram.