How's my Luck now?

Reflections, views and descriptions during my stay at IIM Lucknow from July 2004 to March 2006

Name:
Location: India

Monday, February 07, 2005

'Critical Chain' - II

Some more practical difficulties encountered in project management are as follows:

1. How safety is inserted into the time estimates of almost every step of the project:
It occurs by three mechanisms:
i. Time estimates are based on a past pessimistic experience, the end of the normal distribution curve.
ii. The larger the number of management levels involved, the higher the total estimation, because each level adds its own safety factor.
iii. The estimators also protect their estimation from a global cut, which is expected to come from the top because of external constraints like competitor moves, etc.

2. How most of this safety is actually wasted, ultimately leading to project lateness:
It occurs by three mechanisms:
i. 'Student syndrome' - since the estimator knows about the built-in safety in his estimate, he actually starts late or works less productively, since he knows he has a lot of time.
ii. Multi-tasking - when a single resource is saddled with many things to do at once, and he does a little bit of each task in turn, the lead time of all tasks suffers.
iii. Dependencies between steps - a delay in one step is passed on fully to the next step. An advance made in one step is usually wasted. In the case of parallel steps, the biggest delay is passed on to the next step, while all other early finishes do not count at all.

So, how can these difficulties be solved? With the help of the Genemodem example, Goldratt shows the mechanisms by which the TOC process can be implemented. The changes required are:

1. Identification of constraint:
Realize that the constraint of any project is its critical path.

2. New operational measure to track project progress:
Use 'percentage of critical path already completed' as a measure to track project progress.

3. Project buffer:
All resources should be persuaded to cut their lead time estimates to a 50% completion-chance estimate. Then, to these stripped-down estimates, add a time buffer (the project buffer) at the end of the critical path. Any delay on the critical path would be absorbed by this buffer.
A rule of thumb to decide the size of the buffer: make it half as long as the project lead time.

4. Feeding buffer:
To absorb delays in any step on a non-critical path, a time buffer called the feeding buffer should be added at the point where that path merges with the critical path.
Now, we have a radical recommendation: completion due dates for each step of the project should be eliminated. The resource should just be advised to complete a task as fast as he can.

5. Resource buffer:
One major problem in many projects is resource contention - when one resource is assigned tasks on different paths of a project, and they conflict in time. To remove this, a resource buffer should be implemented. But this is not a delay-absorbing buffer.
It is operationalised in this way: it is "a mechanism of letting a resource know in advance about his/her critical path activity, so that when the time comes to start the activity, he/she leaves everything else and does that activity."

6. Critical chain:
In projects where resource contention is a major problem, the constraint of the project is not the critical path, but the 'critical chain'. This is "the longest chain of dependent steps, where the dependencies may be due to the path, or a result of common resources."
After identifying the critical chain, we can serialize the schedule to remove resource contention, which will involve repositioning of feeding buffers also.

7. Bottleneck buffer:
In case of a resource which is a bottleneck across projects, we have to implement another buffer (a bottleneck buffer) in a way similar to that in which a resource buffer is implemented.

Implementing and adhering to these principles would make projects much more manageable within budget and within deadlines.
In the next post, I will write about a few TOC fundamentals introduced in the book.


1 Comments:

Blogger Greg said...

motivation books what about it..motivation books Here it is now its up to you...

5:48 PM  

Post a Comment

<< Home