
Descriptions and Definitions of Quality Terms, Tools and Techniques 






Six SigmaSigma is the Greek letter used to indicate the Standard Deviation of a distribution. In a Normal distribution, 68% of the population fit within plus or minus one standard deviation of the average. At plus or minus three standard deviations, this goes up to 99.7%, which means that only three in a thousand fall outside these limits. For many quality situations, three in a thousand would be a great error rate, but for many, many others it would be intolerant (think surgical operations, airplane landings or reliability of componentsincluding the multiplying effects of many components in an instrument). So plusorminusthreesigma is not always acceptable. When you get to out to plus or minus six sigma, then the error rate is two in a billion. Or, if you want to allow a little drift of the center of the distribution, then this comes down to around three in a million (3.4, actually), which is a lot tighter tolerance than three in a thousand. To put it another way, three sigma = 1350 ppm (parts per million), whilst six sigma = 0.001 ppm, or 99.9999998% accuracy (or 99.99966% if a drift of plus or minus 1.5 sigma is allowed). So is 'Six Sigma' just a statistical tolerancing technique? Well, yes and no. Mathematically speaking, yes. Business 'fad' speaking, no. Mikel Harry, whilst at Motorola, persuaded them to back a quality improvement program that leveraged the statistical bit and then added a whole lot more. In particular, it involved weeks of training and serious management support to force usage of improvement methods. The trained people are called 'black belts', 'master black belts' and 'green belts'. Thus the real innovation in six sigma is not in the sums or the other tools. It is organizational. It is about leadership, commitment, and all those other things that takes to create real change and make proper use of the tools of quality that have been around for many years. Mikel Harry these days runs the Six Sigma Academy, where their the biggest and most visible customer is General Electric, where Jack Welch's enthusiastic style ensured serious management attention (you didn't get to be a senior manager unless you had done the training and made serious use of it). GE says it saved them around $10B to $15B in the five years up to Welch's retirement. See also:Histogram, Standard Deviation 


