How can I create a project buffer (CCPM-style)?

  • 1
  • Question
  • Updated 10 months ago
Hi,

I am "playing" with the ccpm method.

Has someone an idea how I can create a "project buffer"?

I tried the "normal" buffer in mindjet, but when a task ends earlier and I "cut" all buffers in order to see the new start dates my end milestone moves too.
Photo of Thorsten

Thorsten

  • 18 Posts
  • 3 Reply Likes

Posted 10 months ago

  • 1
Photo of Alex Gooding

Alex Gooding, Champion

  • 872 Posts
  • 220 Reply Likes
If you are talking about incorporating slack or lag time in duration within a task this has been raised a few times but isn't possible in MindManager - see for example this discussion here: https://community.mindjet.com/mindjet/topics/how-can-i-create-a-task-dependency-with-an-offset
Photo of Thorsten

Thorsten

  • 18 Posts
  • 3 Reply Likes
Hi Alex,

thanks for your reply.

I don't mean a slack rather a risk/safety puffer.

In the ccpm method you take the "risk buffers" from all tasks on the critical path and place them in one buffer at the end of the critical path.
Then you reduce this buffer on 50%.

Normal Planning:

Task A+Buffer A -> Task B+Buffer B -> Deadline

CCPM-Planning:

Task A -> Task B -> (Buffer A+Buffer B)*50% -> Deadline

The buffer is flexible:
If a task ends earlier the buffer increase.
If a task needs longer, the buffer decrease.

Then you control, how much of the buffer you used.
0-1/3 (Green): Everthing is ok
1/3-2/3 (Yellow): Everything is under contral but prepare for action
2/3-3/3 (Red): You must act!
Photo of Ary Velstra, Expert Trainer

Ary Velstra, Expert Trainer

  • 1298 Posts
  • 197 Reply Likes
This is an interesting way of planning.

Ofcourse you can add extra tasks .... and call them buffer ... but they are not combined in the way you describe with increase and decrease of the buffers, nor for the totla project.

I am curious how the new Conditional formatting in MM2018 could help.

And ( have been a seasoned projectmanager myself ;-) Why not make a planning that does not need buffers?
If you have enough historic information and use a method like function point analysis like we do in ICT projects, there should be no need for using buffers.
The main reason we want to avoid buffers in our planning, is that teams tend to use it in full - all the time.
Photo of Thorsten

Thorsten

  • 18 Posts
  • 3 Reply Likes
@ Ary: Sorry for the late reply.

I didn't thought that the ccpm method is so unknown (I am newbie in the "structured" project environment ;-) ).
CCPM stands for "critical chain project management" and is an application of the theory of contraints from Goldratt.
The critical chain is the longest chain of tasks after leveling the resources.
Some articles in the www says that it is the critcal path plus resource leveling but after reading the book of Goldratt I think that not the whole truth.

And Ary you are right: the reason for the method of cutting the security buffers from each task and putting them in a summary buffer before the end date is that teams use the whole given time (Parkonson's Law).

The idea of no buffers sounds great. How do you react with Murphy's Law (for example: a supplier delivers later as planned or the duration of a task is longer as planned)?

By the way: What are historic informations? ;-) I am just kidding.
Photo of Ary Velstra, Expert Trainer

Ary Velstra, Expert Trainer

  • 1298 Posts
  • 197 Reply Likes
Hi Thorsten,

Our ICT projects may be a bit more simple than yours...

How do you react with Murphy's Law
(for example: a supplier delivers later as planned

In ICT projects such a supplier knows that this was the last project we give to them.

or the duration of a task is longer as planned)?

This can happen if the planning was not precise enough, or... the developer is not good enough.

In both cases we found that:
It is a rare event in the projects we have done over many years. Repair of such a problem is necessary, but... since it is seldom, this costs less time and less money than -giving that buffer time away-  beforehand.

Photo of Thorsten

Thorsten

  • 18 Posts
  • 3 Reply Likes
Hi Ary,

I would not presume to claim that my projects are more complex.
Often, the solution at the beginning of the project is still a bit vague.
Isn't it the same in ICT? I thought this when I read about all the agile methods.

I read about the Function point Analysis at wikipedia (german version).
It's very interesting.
How do you estimate the point values?
Do you "translate" the point values in man-days for your planning?
Photo of Ary Velstra, Expert Trainer

Ary Velstra, Expert Trainer

  • 1298 Posts
  • 197 Reply Likes
When one calculates Function points it is done by counting fields.
Like how many fields are on a computer screen, or on a print layout etc etc.

Then we use our historic info to translate those points in money - and / or manhours.
The translation factor is a bit diverse based on the type of project.
If you develop a mobile app we may have less history. If we develop a mainframe based -Anwendung-  we have lots of experience so calculate more exact.
We use it also for workflow projects and real-time projects like airtraffic control scheduling etc.

See for lots of info https://nesma.org/
Photo of Thorsten

Thorsten

  • 18 Posts
  • 3 Reply Likes
Hi Ary,

thanks for the informations.