Sorry – it is a German video but it’s quite good to explain, where the term of SIX SIGMA comes from.
A short summary of the video
Our customers have requirements on quality, e.g.
- high technical quality
- high supply reliability
- low rate of complaint
- low error rate
If you supply quantitative measureable quality it shows in most cases a gaussian distribution on its values around a to-be-value and has a committed bottom and top restriction.
The gaussian distribution is characterized by two inflection points. If you cut the gaussian distribution in the middle, at the to-be-value, it results in two s-formed curves, each with one inflection point. At the inflection point of the s-formed curve the direction turns from one into the other. The quality value of this point equals the standard deviation of the distribution. The standard deviation is also called 1σ (ONE SIGMA).
To get an idea whether your quality supply is acceptable or not, your quality distribution should have an error rate within SIX SIGMA – so that the committed restrictions can be achieved in 99.9996% of all cases.