What really is the concept behind 1.5 Sigma shift ? How do we determine whether the sigma level calculated is long term Sigma or Short term sigma level ?

Dear Sandeep,

The 1.5 Sigma shift has been introduced to explain the difference between short term and long term performance. It is hypothesized that over a long period of time things could deteriorate due to poor process control or due to special causes that affects process performance. Hence, a process that is operating at 6 Sigma in the short term will only operate at 4.5 Sigma in the long-term. Thus, a process that is operating at 6 Sigma (in the short term) corresponds to 3.4 DPMO in the long term!

When, people report Sigma levels, more often than not, they report the short term sigma levels (everyone wants to look good!). Practically, all companies use the 1.5 Sigma shift in their reporting and just to be consistent with the rest of the industry, it is recommended that you also use the 1.5 Sigma shift in your calculations.

In conclusion, when you have the long-term data and you compute DPMO numbers and Sigma levels, you get the long-term numbers. Add 1.5 to this number and then report your Sigma level for your process to the rest of the world.

SJ>

Hi SJ,

Thanks a lot for the clarity on the same in simple words.

So that means 6 Sigma (3.4 dpmo ) in ST should be considered equivalent to 4.5 Sigma (Somewhere around 1350 dpmo) in LT

When we say long term, is there any minimal period or timeframe, which we call as long term. Say more than 6 months is long term or it varies from industry to industry ?

What are your thoughts on the same ?

-Sandeep

 

Dear Sandeep,

A 3.4 long-term DPMO corresponds to 4.5 Sigma level in the long term and 6 Sigma level in the short term.

A 1350 long-term DPMO corresponds to 3.0 Sigma level in the long term and 4.5 Sigma level in the short term.

There is no specific timeframe to indicate what is short term and what is long term. It varies from process to process. If a process has only common cause variation for one month but special causes can impact this process month over month then one month could be considered short term. If another process has common cause variation for 1 week but special causes impact it week after week, then one week could be considered short term data. So, you will have to look at your process to determine what is short term and long-term. Ideally, all trends, seasonality, and all types of special causes should be accounted for in the long-term so that long-term DPMO numbers are accurate. Generally, 6 months to 1 year can be considered long term.

SJ.

Dear SJ,

Thanks a lot for bringing the clarity on the same.

Sandeep

 

Dear Sandeep, 

Long Term and Short Term consider time based subgrouping. If another subgrouping approach is used, the equivalent terms used are OVERALL and WITHIN. 

So Long Term is equivalent to OVERALL (considers overall variation)

and Short Term is equivalent to WITHIN (considers within group variation) 

VK

Dear Sandeep

Over the course of time, there may be shift in the mean (which depends on how good of a process control you have) and there may also be an increase in the variation. Both of these factors contribute to the 1.5 Sigma shift.

On any given project, the shift may be more or less than 1.5. It is claimed that historically, across several projects, the shift is about 1.5 between short term and long term.

Whether we like it or not or agree with it or not, we should still use this number so that our calculations (for Sigma level) match those reported by others in the industry as this shift by now is pretty much well accepted.

SJ

Hello all

The Six Sigma learning has been effectively used in various IT verticals like Software Testing, Security, IT Asset Management, domain testing. The programme and techniques also help in small IT ventures. To know ‘how' and ‘how well', read the feedback from industry veterans with Six Sigma knowledge.

Click here to see the feedbacks.

Or paste the following link in your browser:

http://forum.benchma...ion-technology/

Warm regards

AbhishekM