So what is Sigma Shift and what does it mean for you?
Processes tend to behave differently over the short and long terms. For example, they might be stable and predictable over just a few minutes or hours (ie. short term) but highly unstable and unpredictable over several days, weeks or months (ie. longer term).
The difference between the Sigma Levels of a process over the short and long terms is called the Sigma Shift. Historically, people have assumed that the Sigma Shift is 1.5 and have therefore calculated long term capability by collecting only short term data, and subtracting 1.5 from the short term Sigma Level.
This approach misses the point and should only be used as a last resort for estimating long term performance. Instead, if at all possible, you should attempt to collect real data over the long term and then calculate the real Sigma Shift of your process.
Why? Because it provides clues as to how good your process could be if you controlled it better.
A high Sigma Shift suggests that your process could be improved significantly through better control methods.
A low Sigma Shift suggests that your process is well controlled already, and if you want better long term performance, you’ll have to consider a step change in process or technology.