Re[2]: Feature suggestion: 2 times for "duration of task": optimistic and pessimistic

On 22 Aug 2002 13:49:36 +0200 Mikael Hallendal <[email protected]> wrote:

MH> tis 2002-08-20 klockan 01.08 skrev Michele Ravani:
MH> > On Sun, 18 Aug 2002 21:33:50 +0200 info <[email protected]> wrote:
MH> 
MH> > i> When you change the project plan while implementation, you do
MH> loose
MH> > i> oversight about
MH> > i> what was intended originally, what resources/budget were assigned
MH> > i> originally
MH> > i> and who changed where and what.
MH> > i> So it might be a good idea to have a kind of CVS integrated and
MH> be able
MH> > i> to show at what time what changed.
MH> > 
MH> > Versioning, together with baselines, could be a rather powerful
MH> thing.
MH> 
MH> We where thinking of having support to save several baselines. Would
MH> that fill the needs?

Well only partially. Think at a situation were your plan changes because of
additional requirements and similar things. This wouldn't be captured by a
baseline, but would by a versioning system.
On the other hand, baselines could be pushed further. Instead of having a
snapshot (i.e. the last saved baseline) a history could be stored, e.g.
everytime the project is saved (ok, that's versioning) and one could start
producing rather interesting graphs showing the progress vs plan over time
of the project, of a group of tasks, etc. 

MH> > A kind of Value at Risk could be computed, saying something like
MH> 'there is
MH> > a 5% probability that we will overshoot by 30% in time and cost'.
MH> > Definitely and interesting idea.
MH> 
MH> With having pessimistic, assumed and optimistic times on each task this
MH> shouldn't be a problem to do.

Yes, but your are storing (and entering) three values instead of two.
If I say (and enter in the system as a parameter) that from my experience
Joe Coder is within 20% of his estimates most of the times, I can use this
information to get the likelyhood that a task will be delivered in time.
If the work of Joe Coder depends on other tasks, this can be factored into
the computation the same way. Worst case scenarios could be computed.
One could even run MonteCarlo models on such a thing :o)

MH> Thanks for the input. I would very much appreciate if you could all add
MH> things like this in the Bugzilla at (http://bugzilla.codefactory.se)
MH> and
MH> mark them as "Enhancement". This way we won't loose good suggestions
MH> within the mailing list.
MH> 
MH> Regards,
MH>   Mikael Hallendal
MH> 
MH> -- 
MH> Mikael Hallendal                [email protected]
MH> CodeFactory AB                  http://www.codefactory.se/
MH> Office: +46 (0)8 587 583 05     Cell: +46 (0)709 718 918
MH> 


-- 
Michele Ravani                  [email protected]
"Those who live hoping, die singing" My Gran