Six Sigma and Statistical Methods

Six Sigma methodologies use statistical tools used to transform raw data into information. Based on the results, further actions are taken. Statistical tools and related aspects of Six Sigma methodology comprises about half of Six Sigma. In addition, Six Sigma places a lot of emphasis on graphical interpretation of data collected during the course of measurements.

The importance of statistical methods emanate from the fact that many hypotheses can be disproved with sufficient statistical data. The significance of statistical methods in Six Sigma increases with any increase in sample sizes. The statistical methods quantitatively facilitate evaluation of the performance of any process. The purpose of this being to tackle the cost of poor quality (COPQ) first, Six Sigma has a broader scope than the traditional cost effectiveness model.

Some Important Statistical Methods In Six Sigma

Variations in processes are measured in terms of deviation from the mean and data falling within the acceptable statistical limits. Graphical representation of this data helps companies to visualize things with greater accuracy. Let us examine a few of the most commonly used Six Sigma statistical methods.

Control Chart

The deviations within the acceptable limits (upper & lower) are due to common causes. Anything falling beyond the limits is attributable to some specific cause. For example, take the case of writing your name ten times. Although there are similarities, you probably won't be able to find any two signatures being exactly the same. The reason is an inherent variation that produces reasonable results within limits and is termed as 'common cause'. Special causes are those due to forced errors. A control chart has a mathematical mean line in the center and two limit lines. The third component of the Six Sigma control chart is the performance data, which is plotted over time.

You can seek special causes and track common ones through control charts by looking for:

* One set of data falling beyond the acceptable limits (special cause) * Greater than 6 data sets climbing or declining steadily within limits * Eight or more subsequent data sets falling on one side of the mean * Data falling alternately across the mean line

Interpretation emphasizes seeking out the special cause that brings stability to process variation. The real fun here lies in removing the common cause and induced variations also.

Brainstorming & Affinity Group Tool

Brainstorming generates and polishes creative ideas based on the principle that two heads are better than one. An affinity diagram is used to organize & develop brainstorming by fine tuning initial and raw thoughts and removing uncertainties. The advantage of this is that it obviously stimulates for generation of more ideas. The affinity diagram was not originally intended to be a quality management tool. First devised by Kawakita Jiro, the affinity tool emphasizes the need for sorting and titling the data only at the end. A typical affinity diagram organizes the brainstormed ideas on its left panel. On the right side are ideas neatly grouped into affinity sets. Reasons for an idea belonging to a particular set are not given particular importance, but all ideas are clarified. An idea may be present in a single group if it has any resemblance to another.

If there is one thing that summarizes the importance of statistical methods in Six Sigma, it can be none better than a saying, famous in Six Sigma circles - "In God we trust, all the rest bring data". Need we say more?

Tony Jacowski is a quality analyst for The MBA Journal. Aveta Solution's Six Sigma Online offers online six sigma training and certification classes for lean six sigma, black belts, green belts, and yellow belts.