Pert Analysis & Confidence interval

[Jamie Bumsted]

Hi -

Perhaps I am going about this the wrong way, but I would like to utilize
dates with a 95% confidence interval. In order to do this I think I need to
add 2 standard deviations to the finish date after performing the PERT
estimate. I have been able to create a custom column with this duration for
each task, but I can't seem to figure out how to have Project use this
duration to determine the projected start and finish dates. Is there a way
to do this automatically?

Thanks in advance.

Jamie

 

[salgud]

Hi -

Perhaps I am going about this the wrong way, but I would like to utilize
dates with a 95% confidence interval. In order to do this I think I need to
add 2 standard deviations to the finish date after performing the PERT
estimate. I have been able to create a custom column with this duration for
each task, but I can't seem to figure out how to have Project use this
duration to determine the projected start and finish dates. Is there a way
to do this automatically?

Thanks in advance.

Jamie

If you actually want a set confidence level, then PERT is not the way to do
it. It is statistically unsound. You need to look at Monte Carlo analysis,
which is easily done through various software packages.

 

[Jack Dahlgren]

Salgud is correct that PERT is generally discredited as a method of getting
statistically valid dates.
Monte Carlo Simulation is a better approach, but you need to know something
about the process to get good data out.
I have a macro which does monte carlo simulation which you can download and
try for free here:

http://masamiki.com/project/blackjack.htm


-Jack Dahlgre
http://zo-d.com/blog

 

[Jamie Bumsted]

Jack and Salgud -

Thanks for your input, I appreciate you taking the time to answer my
question. I will redirect my efforts to Monte Carlo analysis.

Jamie