Critical Chain Project Management

Basic Features:

a) This technique is based on “Student Syndrome”, which means, start any avtivity as late as possible.

b) While creating estimates, allow only that much time for an activity that there is a 50% probability of completion.

c) Don’t criticize your team members if they are not able to complete the activity in the agreed time frame.

d) Find out the critical path. The chain of activities on critical path will be known as critical chain.

e) Add 50% time buffer in all non critical chains (also known as feeding chains) and 50% time buffer in the critical chain, which will be known as project buffer.

f) If any activity in the feeding chain gets delayed, it will consume the buffer of that particular feeding chain. When the feeding chain buffer is exhausted, it will consume the project buffer.

e) Monitor the feeding chain / project buffer and the activities pending at any time.

Has this technique become popular in industry?



More than 1 critical path reflects bad planning?

Do you think more than 1 critical path reflects bad planning?

The PM books say that more than one critical path will increase risk to the project. My answer is NO. It may even reduce the risk.

Imagine a situation where the committed project duration is 20 days, there is a critical path of 20 days and a non critical path of 16 days in the diagram. In the present planning, if any activity on the critical path gets delayed, then my project will go past the committed date.


So I transfer some resources from the non critical path activities to critical path activities (provided technically it is possible), and in that process make 2 critical paths of 18 days each in the network. Still I am reducing the risk to the project as both the network paths have a buffer of 2 days now as my committed duration is 20 days.

Start to Finish Relationship

The explanation given by PMBOK5 is as follows:

A logical relationship in which a successor activity cannot finish until a predecessor
activity has started.

Example: The first security guard shift (successor) cannot finish until the second
security guard shift (predecessor) starts.

Counter Example:

Imagine you have to do a 2 days of Treatment on the wall which should finish 24 hours before starting painting. Here Painting – although happening later – is the Predecessor (main activity and Treatment is the Successor (Secondary) activity. In this example, the successor activity will finish 24 hours before the Predecessor activity starts. So PMBOK5 explanation is incorrect.

My Interpretation:

In SF relationship, the start of Predecessor (main) activity, DECIDES the finishing of Successor (secondary) activity, with a suitable lead or lag as required.

Does anyone agree with this definition?