How this formulae arrive
3 point estimate and Standard deviation formula:
E = (O + 4L + P)/6 and Sigma(SD) = (P-O)/6
I searched a lot from where these formule are arrived,
Or, you can say how these are derived.
what is the proof.
Have any firm information.
--------------------------------------------
As per my information:
Three point formulae is derived from Simpsons Rule I=3h(f(x0)+4f(x1)+f(x2))
------------------------------------------------------
Where as
SD = (P- O)/6, is nothing but an Difference of -/+ 3 Sigma value at x axis of Normal distribution Bell curve , divided by 6( 3left +3right). or (3SD + 3SD)/6 = SD
Suppose SD (Sigma) = 30, then 3SD = 90, Hence P = +90 from Mean value and O = -90 from mean Value.
SD = (90+90)/6 = 30.
Here a very interesting, condition comes out, In 3 point estimate formula O and P is given. so SD can easily find out and also mean value (Expected estimated days ) E = (O + 4L + P)/6 can be found out.
As per SD formula assumption +/- 3 SD should be Oand P.
--------------------------------------------------------
If I will take Oliver Lehmann Question Here and check it :
| A project manager made 3-point estimates on a critical path and found the following results:
Assuming ±3 sigma precision level for each estimate, what is the standard deviation of the allover path? |
||
| App. 4.2 days | ||
| App. 5.2 days | ||
| App. 6.2 days | ||
| You can not derive the path standard deviation from the information given. | ||
Let I take Activity A - SD = 24-12/6 = 2 , 3Sigma = 6 Now E + 6 = 16+6 = 22 and 16-6 = 10 here O and P are 10 and 22. Not matching with the original value 12 and 24.
Hence this all assumption is failing here.
I would like to challenge Mr. Lehmann, these examples are not fit to the assumption +/- 3 sigma precisions.
Regards
