[Planner] Hello... requests, questions.. Critical Chain processing?
- From: Roberto Leibman <roberto leibman net>
- To: planner lists imendio com
- Subject: [Planner] Hello... requests, questions.. Critical Chain processing?
- Date: Fri, 17 Dec 2004 23:30:41 -0800
Hi,
I've been managing technical projects on and off for a while now, and
everytime I find myself in such a position I start working on my own
project management tool because frankly all out there suck. They suck in
very essential ways; there are fixes and add-ons to some of them which
make them better, but even with that they just don't work. I was even
beta tester for a while for both MS Project and one of the tools on the
market (ProChain).
Have any of you looked at Theory of Constraints type project management?
I believe that it holds a big key into things that would really make
planner stand out from everything else, without the need of making it
overly complex or add every feature under the sun.
I love the simplicity of Planner, and I really think it would be a great
starting point for the kind of thing I have in mind.
The number one mistake that current tools make, according to Critical
Chain thinking is that they all make a very basic assumption that is
VERY wrong: tasks always take a fixed duration (or fixed work). In
reality all tasks are much better modeled as a probability of a task
completing in a given amount of time. Though the Gantt chart (such a
powerful visualization tool) as we know it would have to change
significantly to reflect this change, and as it stands now it is a Bad
Thing (Gantt Chart considered harmful).
Critical Chain project management solves, in a very elegant way, the
whole problem with resource leveling, the self-fulfilling-prophecy task
padding everyone seems to use and I've found that it makes project
management much more focused (you end up looking at very few numbers,
buffers to see how a project is doing), but there aren't any tools to
ease the pain, and shoe-horning current tools is such a bother I end up
using excel (or these days ooffice) to manage.
Here's a good textbook on the subject:
http://www.amazon.com/exec/obidos/tg/detail/-/1574441957/ref=pd_sbs_b_1/002-6900978-8236061?%5Fencoding=UTF8&v=glance
I've found that I can take the leveling algorithms it provides and
manually level (plus calculate and automatically insert buffer tasks) by
hand, it shouldn't be too hard to use it to level resources.
I also like the approach ProChain (a MS Project add-on I used briefly)
takes to update project progress, instead of asking about percentages,
everytime the manager sits to update the project he/she uses a wizard
and updates the number of days remaining on each task currently in the
pipeline (i.e. those not done but whose predecessors are all done),
since new knowledge comes in every day, this number is not always what
you expect, but it is used to update the project in a much more accurate
way.
Anyway, enough of my rant... have these ideas been considered at all?
I'm not doing much management these days, and I only remembered because
I opened up planner for a small project I'm working on alone (which
negates the need for critical chain leveling) and decided to write this
note after a cursory glance to the list archives to see if this has been
addressed.
Perhaps I've picqued some developer's interest, or perhaps next time I'm
managing projects I'll go through the trouble of hacking into Planner to
get it to do what I want.
What do you think?
Roberto
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]