need help to calculate sixsigma

This topic has 7 replies, 6 voices, and was last updated 14 years, 12 months ago by Sudhi.

Viewing 8 posts - 1 through 8 (of 8 total)

Author

Posts
February 7, 2005 at 4:36 am #26253

Vikas
★ 10 Years ★
Member

@Vikas Include @Vikas in your post and this person will
be notified via email.

Sir;
I am trying to calculate six sigma by using following formula:
sixSigma=0.8406+SQRT(29.37-(2.221*(1/DPMO)))
WHERE DPMO=(611/1081)*1000000.
The answer comes is sixSigma=6.26 (which is wrong).
Please explain me what is wrong in my formula.
(0.8406,29.37,2.221 all I taking as a constant)
regards
vikas

0

February 9, 2005 at 5:38 pm #63793

Tajjammul Hussain
★ 10 Years ★
Member

@Tajjammul-Hussain Include @Tajjammul-Hussain in your post and this person will
be notified via email.

Vikas,
Why are you complicating the calculation. You can either:
Use the DPMO to Sigma Conversion Table
OR
If you plan to do this on an Excel, then maintain a master table with sigma and DPMO values and use a Vlookup Function to give you the sigma value.
Hussain

0

February 10, 2005 at 4:16 am #63794

Vikas
★ 10 Years ★
Member

@Vikas Include @Vikas in your post and this person will
be notified via email.

Dear Sir
thanks for reply.
I dont want to calculate sigma in Excel.
what is the formula to calculate sigma using dpmo?
regards
vikas

0

February 10, 2005 at 11:34 am #63795

Tajjammul Hussain
★ 10 Years ★
Member

@Tajjammul-Hussain Include @Tajjammul-Hussain in your post and this person will
be notified via email.

Ok…You just need to go one step back and look at yeild % and then calculate sigma using the following :
= Normsinv(yeild%) + 1.5 = Short Term Sigma

0

February 16, 2005 at 1:28 pm #63799

JC
★ 10 Years ★
Participant

@JC Include @JC in your post and this person will
be notified via email.

Vikas,
Where did you get your fomula? It looks like one of your colleagues regressed a portion of the Z table. Assuming your data are long term, normally distributed, etc., I calculate a benchmark Sigma of 6.37 (let’s get real: 6.4). You calculated 6.3. Not bad actually. Using the Z table or Excell, I can get a more accurate estimate than you do, but do I have better information than you? Not really. You already have your yield.
If you have a few processes to benchmark, and you don’t have a Z table, try the Sigma Calculator on iSixSigma site. If you need to crunch more (for a Monte Carlo analysis?) go high tech and use Excell.
Good luck.

0

February 16, 2005 at 3:49 pm #63801

Slava
★ 10 Years ★
Member

@Slava Include @Slava in your post and this person will
be notified via email.

You have forgotten to put a “ln” in your equation.
process Sigma = 0.8406 + SQRT(29.37 – 2.221 * (ln(DPMO)))
you can check it here – https://www.isixsigma.com/library/content/c020507a.asp

0

February 25, 2005 at 8:33 am #63808

CB
★ 10 Years ★
Participant

@CB Include @CB in your post and this person will
be notified via email.

Calculate DPO first by using this formula:
DPO = Total no. of defects / Total no. of units * no. of opportunities per unit
Then calculate DPMO = DPO * 1000000.
Next step is to look at the corresponding table and find out the sigma. Short term and Long term.
Do let me know if you need the table.
All the best.

0

August 20, 2007 at 1:37 pm #64816

Sudhi
★ 10 Years ★
Member

@Sudhi Include @Sudhi in your post and this person will
be notified via email.

The formula being used is valid as long as the DPMO is <550000 (i.e <55% defects) which would be the case in most situations.
In your case as mentioned it exceeds the limit (601/1081 = 56%)
So it is inappropriate to use this formula

0
Author

Posts