The critical chain theory for project management was developed by Eliyahu Goldratt and presented in a rather novel manner, through the form of a literal novel, Critical Chain. The focus of critical chain is to improve on what are seen as some of the core flaws of standard project management task estimation. The biggest of these standards that critical chain takes issue with are:
- Overestimation of task length: Adding pad or using float, these additions are then compounded by each level of estimation inputted
- Student syndrome: Knowing that you have plenty of time in which to complete the task, you don’t start it until the last minute
- Parkinson’s Law: States that “work expands so as to fill the time available for its completion.”
Key Suggestions of Critical Chain Theory
Goldratt set out to make projects more efficient in terms of estimation accuracy, resource usage and time control. This is to be achieved through the following alterations.
Buffers Instead of “Pad” and “Float”
The critical chain theory adapts the concept of “pad” (i.e., the extra bit one adds on to an estimate to give it a better chance of completion) and “float” used in critical path method to describe the “delay space” a project has. The alternative buffers proposed were:
- Project buffer: This is basically a collection of all the time that would otherwise be padding or float. It is added together and placed at the end of the critical chain, between the final task and project close. It then becomes like a centralized resource, with the project manager or team leads responsible for deciding when and how much of it can be used.
- Feed buffer: To ensure that non-critical tasks on the chain don’t end up delaying critical tasks, a feeding buffer is created at the end of the non-critical chain to ensure it has the space to cover any unexpected delays.
- Resource buffer: This is a brief interlude before a critical human or other resource is brought into action on a task. It thus serves as an indicator to ready that resource for the new deployment.
50% Success Estimation
Another important suggestion included in the critical chain is that of using two estimates, one that has a 50% chance of success and another that gives a “safe” estimate that has an 80% or 90% chance of success. The concept behind this is that the 50% estimates will even out over the course of the project between success and failure, supposedly leaving 0% variance on project delays, while the “safe” options unnecessarily build delays into the task. If there is too much one way or the other, it means that the estimation process itself is off.
Is critical chain theory still relevant for modern projects?
While the core ideas of critical chain theory concerning the excess padding that can be trimmed from project estimates do hold some weight, it lacks on providing alternatives to errant scheduling. In addition, some believe the idea of buffers and who decides when to use them adds an unnecessary layer of complexity and oversight which may only serve to slow down Agile decision-making.
In any case, modern project management software, such as the Planview AdaptiveWork suite of cloud-based programs, creates a system that offers instant insight and greater control over resource usage and task scheduling. Learn more with a live demo.
