Critical-Chain Project Management

Introduction

Projects are often managed under tight schedules and limited resources, with the managers sweating everyone to accomplish set objectives within the specified timeline. One of the most important aspects of project management is effectively utilizing the available resource to achieve the project’s goals. In the past, project managers used different models when scheduling and monitoring the different aspects of projects. Some of those models include the Critical-Path Model (CPM), the waterfall project management model, the DMAIC model, etc. In particular, the critical path model has been favored because it can estimate the throughput of the project, which is essential for planning all the necessary resources and making project-related commitments. However, critics often highlighted that the critical-path model tends to overlook the effect of resource-related constraints on a project’s completion schedule.

The critical-chain project management model (CCPM) developed from Eliyahu Goldratt’s book, The Goal, in which he narrated the experiences of a plant manager at UniCo’s manufacturing plants (Goldratt & Cox, 2004). In his book, Goldratt explored some of the common stereotypes of effective project management. The book is centered on Alex Rogo, who was appointed to steer the plant from its loss-plagued operations. However, despite having a degree and masters in engineering, Rogo’s plant continues to turn in losses, with the majority of its customer feeling unsatisfied with the ever-late shipping of their products. Rogo eventually meets up with his college lecturer, who challenges him to re-evaluate his pre-understandings on various aspects of effective management. In the end, Rogo develops the Theory of Constraints (TOCs) that enables him to lead his plant from the poorest performing plant in the company to the most efficient and profitable.

Goldratt later developed his theory of constraints into the CCPM model in his 1997 book, Critical Chain (Meredith et al., 2016, p.216). Therefore, the CCPM model is scheduling technique based on the theory of constraints that analyzes tasks dependencies, resource constraints, and buffer requirements in project scheduling. This paper will review Goldratt’s CCPM model, analyze its pros and challenges, and compare the CCPM to traditional project management scheduling techniques.

The Basis of the CCPM Concept

In any project, the lead time is always an important factor. The client, the organization, and the project managers are always concerned with the duration a project would take because such information helps in the planning of resources. In most cases, project managers make estimations of the project’s lead time by summing up the estimated time necessary to complete each stage. For instance, in a part production process, the manager would need to consult each station operator to know the time required to complete any given production process. However, Larson & Gray (2018, p.295) argue that most operators tend to overestimate their time limit to increase their chances of completing the task ahead of schedule. Consequently, the project manager is likely to add more time to the project’s lead time summation as a safety precaution on the manager’s part. Nevertheless, Larson & Gray (2018) still question why most projects often end up finishing behind their schedule despite the overestimated schedules at their commencement.

Delays in project completion often result from two factors, which are poor time estimations techniques and poor behavior among employees. Estimations of project lead times are mostly based on the manager’s and other stakeholders’ presumptions of the project’s activity path. The activity path refers to the processes required to achieve the end-product. However, in most cases, the activity path may consist of various new or unpredictable processes, which create a lot of uncertainties in each process’s actual duration. Consequently, it becomes difficult to determine the exact duration that each process in the critical chain will take. Therefore, if the project manager or station operator wants to impress the client, they are likely to underestimate the project’s schedule, which will lead to project delay. Conversely, if the manager or station operator desires to protect themselves from delay liabilities, they are likely to overestimate the project’s schedule, which could result in poor work ethic among the employees.

The second reason that causes lateness in most project schedules is poor work ethics or unprofessional behavior. Shurrab (2015) argues that several unethical employee behaviors are likely to delay the completion of a project despite a precise estimation of the project’s lead time. Such behavior includes student syndrome, Parkinson’s Law effect, deliberate padding, dropped baton tendencies, and excessive multitasking (Shurrab, 2015, p.1; Larson & Gray, 2018, p.295). Goldratt argued that students often procrastinate or delay starting their term papers until the due date is imminent. He referred to this behavior as the student syndrome. He also argued that employees are likely to procrastinate the starting of project procedures when they feel that they have more than enough time. According to Larson & Gray (2018, p.296), such procrastinations compromise the worker’s opportunity to identify and resolve obstacles within the allocated time.

Parkinson’s Law asserts that the magnitude of the work required fills the allocated time for the project (Larson & Gray, 2018). Goldratt argued that workers are likely to use the extra allocated time for the project to carry out other unrelated activities. Therefore, he claimed that the “perceived” free time in the project’s schedule tends to mislead the workers into procrastination, eventually resulting in delayed completion dates. Also, Goldratt argued against multitasking during project processing. According to Shurrab (2015), bad multitasking refers to working on a new project before completing the previous one. Such behavior causes chaos within the critical chain that eventually leads to delayed completion timelines.

Additionally, deliberate padding refers to self-protective behavior among workers where they slow down or do other unnecessary processes to prolong their lead time. Larson & Gray (2018) argue that workers are less likely to report early process completions because they want to keep future expectations at “manageable” levels. For instance, if a machinist had estimated that the process would take five days, they may fear that handing in the completed part after two would jeopardize the estimation timelines for the next project. Therefore, such an employee may choose to slow down the machining to make sure the process takes a couple more days. However, such behavior causes a higher depletion rate of the whole project’s buffer time, leading to possible delays in completion at the end of the critical chain.

Another possible cause of project delays is what Larson & Gray (2018) refer to as “dropped baton” within the critical chain processes (p.295). Goldratt argued that the critical chain was similar to a relay race, where one station receives a baton that they then hand over to the third party after completing their race. However, the author argued that time is lost whenever members in one stage of the critical chain fail to inform the succeeding group of the completed process. For instance, an employee may finish his or her process and proceed to other activities without informing the next group of the project’s status. In some cases, the project may lie ideal for several days before someone bothers to follow up on its progress. Therefore, the project manager must be keen to eliminate “dropped baton” time losses within the project’s critical chain. Goldratt suggested that eliminating such delays due to “dropped batons” require the adoption of daily reporting and constant information sharing between adjacent stations within the critical chain. Such activities will enable different station members to adequately prepare for the project, which will enhance the seamless flow of work through the chain.

The Cost of Overestimated Project Schedules

Overestimation of project schedules can lead to a couple of problems for the organization. First, this could lead to poor performances and inefficiencies among the workers. According to Meredith et al. (2016, p.219), allocating too much time to projects could lead to unethical workers’ behavior, such as the student syndrome, deliberate padding, or baton dropping. Workers adopt such behavior to protect the time allowances of future tasks. However, such behavior puts more strain on the organization’s performance and efficiency because it consumes unnecessary resources.

Additionally, the overestimation of the project’s schedule could also result in additional costs or penalties. Meredith et al. (2016, p.219) argue that the early completion of a project will need additional resources for storage or maintenance before the commencement of the next phase in the critical chain. Such storage and maintenance resources may result in additional costs, which further depreciate the company’s performance. Goldratt’s CCPM model addresses all the aforementioned challenges of poor and overestimation of project lead time. The CCPM differs from other scheduling techniques because it addresses the uncertainty of limited resources as well as the dependencies between events within the critical chain.

The Theory of CCPM

According to Larson & Gray (2018, p. 295), the critical chain of any project is the “longest string of” dependent events, which refers to all the mandatory preceding activities before the project’s completion. In this regard, the CCPM model aims at reducing or eliminating delays within the critical chain due to uncertainty, overestimation, or “wasted internal buffers” (Shurrab, 2015). Similar to the critical-path model, the CCPM adds protection to each process within the critical chain by allocating time allowances, otherwise referred to as buffers (Baldwin & Bordoli, 2014). However, the CCPM differs from the CPM by aggregating these buffers at the end of the critical chain and refining the aggregate buffer into an effective project scheduling and monitoring tool.

The Model in Practice

The CCPM model aims at eliminating any exaggerated overestimations of the project’s lead time. In this regard, Goldratt emphasized the importance of using “true 50/50” time estimates when developing the project’s schedule (Larson & Gray, 2018, p.296). In traditional methods, project managers tended to use safe predictions of time estimates. For instance, if a machinist is asked the duration that carrying out a specified procedure, he would have an exact estimate that he believes is enough. Goldratt argued that such an exact estimate represents the “true” time approximation that gives the machinist a fifty percent chance of achieving the task. However, he also often observed that most workers prefer commitments that offer a higher probability of success. Goldratt argued that using the “true” time estimate eliminated poor work ethic among employees, which results in resource and time wastage.

The first step in the CCPM method is to collect all the time estimates for each individual task within the critical chain. To collect these estimates, the project manager must consult with each member of the critical chain. Then, the manager should revise these time estimates into true 50/50 approximations. It is assumed that the majority of workers present time estimates with between 80 and 90 percent assurance. To carry out such revisions, the project manager must also enquire about the least time estimates for completing each individual task. The CCPM assumes that the least time estimates represent the actual lead time given that the process runs uninterrupted with full devotion of the worker’s efforts (Baldwin & Bordoli, 2014, p.153).

The next procedure in the CCPM involves drafting the time estimate on a logic diagram, which enables the project manager to accumulate the project’s lead time. The most common logic diagram for drafting such time estimates is the Gantt chart. First, the time estimates should be drafted with considerations of both resource dependencies and precedence along the critical chain (Meredith et al., 2016, p.224). It means that the manager should draft the time estimates based on the latest starting time. For instance, if the critical chain requires a resource A, which is in current use, the starting time for that task should be at the maximum possible end of the current task. In the project’s critical chain, the initial draft should include the overestimated time limits for each individual task.

The next step involves compressing the schedule to eliminate the “safety” included in each individual task. The step requires the manager to re-calculate the “true” time estimates based on the idea that the initial estimates represent an 80-90 percent probability of completion. However, in re-drafting the logic diagram, the critical chain schedules must maintain the starting times from the previous diagram. Later, the manager develops the cumulative “true” time estimate of the whole critical chain by cumulating all the individual “true” time estimates. The manager will then sum up all the overestimated time limits. Finally, he will re-draft the Gantt chart after adding a percentage of the aggregate time overestimates, as a project buffer. Therefore, the project buffer replaces the series of individual safety time into a single safety duration for the whole critical chain schedule. According to Larson & Gray (2018, p.296), the recommended percentage for the project buffer is fifty percent of the aggregated time overestimates. Figure 1 shows an example of a revised critical path based on the CCPM model.

Additionally, the CCPM model suggests the addition of two more buffers along the critical chain, which include feeder and resource buffers. Feeder buffers refer to additional noncritical paths that merge with the critical path at the point of constraint, also known as the bottleneck. The bottleneck represents the task or station with the least processing capacity throughout the critical chain. Feeder buffers are purposefully added to protect the bottleneck from any delays or stoppages due to resource or material unavailability. The CCPM also suggests adding resource buffers that act a time buffers to caution the critical chain against delays or stoppages due to resource unavailability. Such resource buffers may include storage resources that ensure scarce resources are always available on demand. Other resource buffers may include redundant processes that increase the availability of scarce materials. Both feeder and resource buffers provide further protection to the critical chain lead time.

Figure 1: A revised critical chain schedule based on CCPM model (Baldwin & Bordoli, 2014, p.154)

Critical-Chain Monitoring

The allocated project buffer plays a vital role in monitoring and measuring the project’s progress towards completion. The project buffer consumption rate along the critical chain acts as an indicator of the schedule being behind or at par with the estimated completion date. High consumption rate indicates that the project is underperforming, while low consumption rates suggest that the project will be completed within the estimated period.

Conclusion

The CCPM model provides an elaborate method of scheduling the critical chain of any project, determining the project’s lead time, and monitoring its progress. The method addresses the uncertainty brought about by both strained resources and activities. However, the main advantage of the CCPM model is that it focuses on the critical chain’s bottleneck by managing the availability of feeder and resource buffers, which minimizes delays due to unnecessary process stoppages. Secondly, the method reduces unnecessary time allowances, which tend to promote unethical work ethics resulting in poor resource and time management. Additionally, the methods buffer management strategies act as vital tools for monitoring the progress of the project. By constantly monitoring the project’s progress, the project managers can make informed, proactive decisions that improve both the coordination and the performance of the critical chain.

Critics, however, argue that the CCPM model fails to address other project objectives, which may be similarly important compared to the lead time. Such objectives may include high quality, process costs, scope, etc. Critics may also argue that the CCPM model can only be useful when scheduling renewable resources, such as machine use, labor, etc.

References

Baldwin, A., & Bordoli, D. (2014). Critical Chain Project Management. A Handbook For Construction Planning And Scheduling, 151-157. https://doi.org/10.1002/9781118838167.ch5

Goldratt, E., & Cox, J. (2004). The Goal (3rd ed.). North River Press.

Larson, E., & Gray, C. (2018). Project management: The Managerial Process (7th ed., pp. 294-303). McGraw-Hill education.

Meredith, J., Shafer, S., & Mantel, S. (2016). Project Management In Practice (6th ed., pp. 216-232). Wiley.

Shurrab, M. (2015). Traditional Critical Path Method versus Critical Chain Project Management: A Comparative View. International Journal Of Economics & Management Sciences, 04(09). https://doi.org/10.4172/2162-6359.1000292