Sigma, the word which forms the core of this process is derived from the Greek alphabet and is used to represent the standard deviation of a statistical population.
The term Six Sigma process arrives from the belief that if one has six standard deviations between the standard specification limit and the mean, there will be a few or no item that fails to meet the specifications. This study is based on the calculation model which is used in a process capability study where the number of standard deviations that occur between the mean and the nearest specification is in sigma units.
What happens here is that once the process standard deviation rises, fewer standard deviations will find its place between the nearest specification limit and the mean, which in turn will decrease the sigma number.
It has been studied that processes do not perform in the long run, thus, the number of sigmas that fit between the processes mean and the nearest specification limit might drop over time.
It is to cater to this real-life increase in variation over time that an empirically based 1.5 sigma shift was introduced in the six sigma calculation. It is from this idea that the process that fits the process mean and the nearest specification limit in a short-term study will in the long run fit 4.5 sigmas. This study finally gives rise to the widely accepted definition of the Six Sigma process, which produces 3.4 defective parts out of per million opportunities. To sum up, we can say that the Six Sigma process corresponds to 4.5 sigmas, which is namely 6 sigmas minus the 1.5 sigma shift which is introduced in order to account for the long-term variation.