Critical Chain Method in Project Scheduling

From apppm
Jump to: navigation, search

Developed by Georgios Nikou

The Critical Chain Method (CCM) is a relatively new concept in Project Manager, aiming to manage and control both projects and portfolios. By focusing on the availability of resources, while also promoting a Relay Race culture in mentality, it increases productivity and performance. The method is based on the Theory of Constraints (TOC), suggested by Dr. Eliyahu Goldratt and since then, its application has been very successful in a large number of actors. Its main difference from traditional Project Management techniques is that excess multitasking is avoided and focus is shifted to resources and early finishes instead of specific deadlines and strict order enforcement.


Methodology Background

The theory of constraints, which is the foundation for the Critical Chain Method, was first introduced by Dr. Eliyahu M. Goldratt. Having a rich background as an educator, author, scientist, philosopher and business leader, Dr. Goldratt was an unconventional and successful educator with a charismatic tendency to provoke his listeners to think and a large number of successful companies and entrepreneurs at the time were seeking his advice.[1] In his book name “The Goal”, written in 1984, he describes the challenges that are faced in a manufacturing plant. Dr. Goldratt suggests that a system capacity and efficiency are based in its elements, as well as their interdependencies. That is where a contradiction is observed. For the constraints to be more efficient and increase their productivity, the rule “more is better” is applied. However, when talking about a non-constraint element, the rule “more is worse” is followed, since if its productivity was increased, there would be an excess of resources, choking the system as a whole. Following the above observations, Dr. Goldratt states that three measures are able to control the aforementioned plant and in the following order: Throughput, Operating Expense and Inventory. The Drum-Buffer-Rope (DBR) Method is also introduced, in order to ensure that the plant will produce only what is necessary, avoiding wastes in overproduction. This methodology was named Optimized Production Technology and it was later evolved into the Theory of Constraints (TOC) and in Dr. Goldratt’s third book, named “Critical Chain” (1997), it was introduced into Project Management.[2]

Theory of Constraints

A distinct trademark of this theory is the optimal prioritization of improvement activities, leading to rapid improvement when there is urgent need for change. The benefits obtained from applying the Theory of constraints are:

1. Increased profit

2. Fast improvement

3. Improved capacity

4. Reduced lead times

5. Reduced inventory

In its core concept, the Theory of constraints uses a cycle with the following steps:[3]

1. Identify the constraint

2. Exploit the constraint

3. Subordinate and synchronize to the constraint

4. Elevate the performance of the constraint

5. Repeat the process

Methodology basis

The Critical Chain Method puts a great emphasis on the human nature during Project Management and provides an insight on what may happen when this discipline is applied to real persons.

Estimating: When someone is asked to estimate the duration for a given task, his first estimation will usually be somewhat ambitious. After reconsidering the task at hand, that person will ask for a longer duration to deliver, providing himself with a personal failsafe, in case of unpredicted challenges that might occur. This safety, created to avoid negative attention, while not inherently wrong, is a hidden delay in the project.

Student Syndrome: Following an example in Dr. Goldratt’s book “Critical Chain”, a group of students is given an assignment from their professor. At first, the students think that the assignment requires more time and ask for a longer deadline and their professor agrees. Reflecting back on the available time to work, the students consider that they have more than enough time to complete their assignment, so they decide to begin later, putting off the start of their work until the last minute. However, due to challenges that might occur when working on the task, they realize that they are running out of time, so they spend the last days available working overtime, overrunning their initial estimate. It is therefore observed that most people fall into the student syndrome when working on a task with a hidden safety margin.

Parkinson’s law: Work is always organized and shaped in order to fit the time allotted. When a task has a strict deadline, it is very rare that it will be completed before this time limit. This adjustment comes in a multitude of ways and the persons that are engaged in the task will try and modify their work rate accordingly, in order to deliver exactly on time. Traditional Project Management techniques stress the importance of not being late, while they fail to provide incentives for early finishes. Thus, hidden safety, Student syndrome and Parkinson’s Law effects are encouraged in the working environment.

Multitasking: Multi-project environments are dominant nowadays and multitasking is very commonly observed, having resources and personnel working on a task, switch in the middle of their work, in order to resume later. This organize chaos stems mainly from the fact that the majority of clients believe that their demands are of the highest priority, thus putting pressure on project managers to show progress towards each one of them. An impactful change that is required is to prioritize which tasks are most important for the organization and complete them first, before moving on to the rest. Switching from micro-level to macro-level decisions, a better performance is achieved and less time is spent in constantly switching the allocation of resources. This way, the most important tasks/projects are finished early, optimizing the system’s ability to perform and deliver.

No early finishes: Having already talked about the Student syndrome and Parkinson’s Law, there is an additional factor preventing early finishes in projects. A person that finishes his task early risks being accused by his colleagues for the way he works and very rarely is he rewarded for his performance. Another risk this person might encounter is that future estimates concerning his work will probably be cut, due to his previous early finishes. Concluding, another challenge that might arise is that despite that person’s early finish, the project will be unable to continue, due to the fact that the resources required are allocated in another task, simply because it was unexpected to have an early finish, thus an early task on to the next step.[4]

All the aforementioned phenomena showcase that when traditional Project Management techniques were used, the best a Project Manager could expect was to finish exactly on time, observing many delays quite often. By introducing incentives for early finishes, and decreasing multi-tasking, Dr. Goldratt suggested the Critical Chain Method as an improvement. It is worth noting down that Dr. Goldratt did not suggest to stop multitasking, as that would be simply inevitable, taking into account the current market state. What he proposed was to apply it to a lesser extent, essentially removing what he called “Bad multitasking”.[5]

Introducing TOC to Project Management

This section analyzes the steps that were previously mentioned when the Theory of constraints is applied to Project Management.

A. Identify the constraint

The theory identifies the Critical Chain as “The sequence of dependent events that prevents the project from completing in a shorter interval. Resource dependencies determine the critical chain as much as do task dependencies.” The definition of the constraint is implemented form the impact that schedule has on project cost and project scope. These conditions are always as dependent as for example, a bigger scope in a project is bound to increase its cost and prolong its schedule required. The Critical Path method that was previously used did not take into account the resources allocated, therefore it can be implied that if those resources were infinite, the two methods would be exactly identical.

B. Exploit the constraints

In this step, the activities on the Critical Chain are scrutinized, in order to ensure no delays and eliminate the factors that might cause them.

The first step is to exploit the estimates in task duration. Usually, these estimates are calculated with a probability of 90% completion. However, Critical Chain Method suggests estimates with a 50% probability, entirely removing all safety margins and considering that the project will be completed without any challenges arising during its process.

The second step is to exploit the statistical law of aggregation. This statistical method that combines variances is able to protect a chain of activities with the same level of probability, instead of protecting its individual activity separately, using considerably less contingency time. This is done by introducing buffers to the end of the path. These buffers are activities in the chain that require no work and act as a failsafe, should a delay occurs. The three types of buffers that are used are as follows:[6]

Project Buffer: Placed between the last task and the project’s completion, this non-activity buffer acts as a contingency, aiming to protect the scheduled deadline. Any deadlines in the Critical Chain will consume this buffer and any early finishes from other activities will be added in it. Its usual duration is 50% of the contingency that was removed from all task estimates.

Feeding Buffer: these buffers are added to the non-critical chain, in order to ensure that any delays there will not affect the critical chain. They are calculated in the same way as the Project Buffer, based on a fraction of the safety removed from the non-critical tasks.

Resource Buffer: They are kept along with the critical chain in order to ensure resource availability.[7]

C. Subordinating the constraints

As it was previously stated, the critical chain must be protected from non-critical activities. This is done by introducing Feeding Buffers at the end of non-critical chains that feed into the critical chain. As mentioned above, these Feeding Buffers are calculated as an aggregate, exploiting the statistical method introduced in step B. This control mechanism is therefore able to subordinate paths that feed into the critical chain and ensure that non-core delays will not transfer into the core activities.

D. Elevate the constraints

When the improvement is maximized, more resources should be used in order to increase the system’s efficiency and capacity for production. As it was previously mentioned, the main challenges that Critical Chain Method aims to face are excessive multitasking and the lack of incentives for early finish. For these reasons, there are no fixed dates, except for the end of the project (delivery) and the critical chain activities. Additionally, resources are ready and available whenever a task is finished and requires them and works flows uninterrupted, as long as a critical task is concerned.[8]

To point this out, in many cases the staff that is working on a critical activity is given a kind of symbol (such as a red flag) to indicate that they should stay uninterrupted and focus on their task at hand. Concluding, Critical Chain method promotes a “Relay Race” culture during which, every team stands ready to continue immediately when the previous team has finished. Thus, focus remains on winning the race and constantly improving the working rate.[9]

Ε. When the constraint has been improved, go back to step A and repeat the process

Principles and application

The method is known to enforce predictability levels to up to 90%, while also increasing productivity and reduces speed by up to 30%. This is achieved by reducing errors and wasted time, leading to reduced stress while working on the project. An additional measure to be taken into account is shifting from deadline management to resource management. This way, the manager responsible should not focus on the deadlines that are to be met but on the availability of resources, making sure that every task is able to start, as soon as the one that was preceding it ends.

Scheduling globally

The first step is to apply urgency, both in single-project level (individual tasks), as well as in portfolio-level (individual projects). After each task has been scheduled and the network of activities is finished, priorities are set. The longest chain of activities is therefore called the “critical chain”. Once the critical chain has been identified, task estimates are reduced and they are pooled together into the project buffer, which is placed before the end of the project, in order to protect its deadline. Managers use the fever chart, in order to measure the amount of the buffer that has remained in relation to the percentage of the project that is completed. If they find themselves in the green zone, everything is proceeding as normal. The yellow zone indicates that the situation must be monitored, but it is still in control. The red zone is a warning that corrections must be made, else the project risks exceeding the time that was requested.

Priorities instead of excess multitasking

CCM promotes the reduction of multitasking, applying it only when it is absolutely necessary. By maintaining focus on important tasks, less time is wasted by shifting between them and stress is also reduced. When two tasks/projects need to run at the same time, the method suggests that one of them is finished first, before moving on to the next one. This choice is based on the importance of the tasks/projects at hand and the limitations imposed by the resources that are available. The critical chain is protected by the feeding buffers inserted before the end of non-critical tasks, in order to ensure the overall performance of the system.

Relay Race culture

In order to change the culture from being focused on specific deadlines, the organization employing the Critical Chain Method should create a Senior Leadership team, overseeing the shift in the working environment, rather than the implementation of work. Additionally, a rules change is needed to prevent employees that works in a critical from being called onto other matters. A simple measure is to hand in the right to say “no” or “not now”, even to their seniors, until their work is done. Finally, more communication is required in order for these changes to take place. It is therefore suggested that all parties meet constantly in order to exchange information and feedback on how their performance can be increased and discuss the results.[10]


The following video, found in Youtube, showcases a simple example of how CCM can be applied to projects:[11]

Critical Chain Example

As the video suggests, the first step when applying the method is to create the schedule, taking into account dependencies and durations. The next step is to add buffers into the project and calculate them accordingly. When the buffers have been added, the critical path is identified.

Feeder buffers have to be added after the of each critical path. Once this is done, all dependencies should be once again checked and updated. The task estimates are reduced according to a probability of success at 50%. It is worth noting that if the feeders buffers that are added create a new critical path, they should then be reduced even more, up to the point of not creating that new critical path.


One great advantage that CCM offers is the fact that the manager is in control of all contingency time and safety factors and has the freedom to allocate time accordingly. The fever charts that show the buffer consumption are able to provide information on the project’s process, thus enhancing control over task and resource management. This way, it is significantly more possible to deliver earlier on schedule, rather by maintaining a deadline focused mentality. In addition, a large number of successful companies has observed great improvement in their performance by employing the Critical Chain Method, despite the fact that it is still a new practice, increasing their benefit and productivity levels.

Critical Thoughts/Limitations

One issue that raises doubts about the Critical Chain Method (CCM) is the fact that its theory suggests that task owners will always overestimate the duration required and then they will try and allocate their effort accordingly, in order to meet the requested deadline. A study by Hill, Thomas and Allen in 2000 [11] showed that 60% of the tasks studied had finished earlier. While the safety factor of 95% was definitely not met, it shows that the tendency to exceed the deadline was not so strong, as CCM suggests.

Another issue is the fact that CCM is unable to provide exact calculations when estimating the safety factor inserted in each task. Due to the variety of behavioral aspects built into scheduling, it would be very hard to obtain a formula that simulates the correct amount of contingency, however, CCM only deals superficially with these aspects. In addition to that, there is always the risk that project managers will not be willing to agree in shortening their safety factors and merging them into buffers, due to the different conditions that are faced in each individual project.

The estimation of buffers also proves to be another arguing point. CCM suggests that in order to be able to deliver in a shorter period of time, thus improving scheduling, the buffers that were consumed have to be less than the safety margins removed from task estimates. The chain structure suggested might be true when dealing with projects in construction, manufacture and integration tasks. However, projects involving a core of activities as a starting point, such as analysis and design, lead to complex systems, where the allocation of the buffers inserted is not clear.

Multitasking is also seen as a problem in CCM and the method suggests avoiding it as much as possible, maintaining focus on the most important tasks first and foremost. However, there are also studies that suggest multitasking is better for the overall performance of company/system. This means that until more research takes place and solid results are obtained, one is not able to exactly state the correct amount of multitasking that is required for improved performance.[12]




Annotated Bibliography

1. This website contains a detailed analysis as an introduction to the basics of the Theory of Constraints, providing a very useful insight into the development of this tool in Project Management

2. A detailed analysis on the Critical Chain Method, focusing on the needs that led to its development and the challenges it aims to face when used in Project, Program and Portfolio scheduling

3. A paper that provides an analysis on the steps included in the Critical Chain Method and on how they are applied in Project management and scheduling

4. A video on Youtube that provided an excellent introduction to the Critical Chain Method, highlighting its advantages over the more traditional techniques in Project Management

5. A very useful example found in a video on Youtube, explaining how the Critical Chain Method can be applied in a small, proxy project and what factors should be taken into account during scheduling

6. A Project Management blog, thoroughly explaining the different kind of buffers used in the Critical Chain Method

7. This paper analyzes the uses of Critical Chain Method in Project Management and is focused on a critical approach, highlighting the limitations of this particular technique when applied

Personal tools