By Brian J. Furniss & Theodore J. Trauner
Most everyone in the construction industry says he or she understands what the critical path of a project schedule is and what it means. The majority of the time, the understood definition of the critical path is correct, but that definition is often based on traditional scheduling methods. As computers and scheduling software have become more common, anyone can create a schedule, click a few buttons, and get a printout of the critical path. However, the evolution of project scheduling requires that schedulers and project managers gain a much more thorough understanding of the definition of the critical path. Because today’s software packages are more robust and allow us so many more options to schedule our project more accurately, the art of scheduling has evolved tremendously from when schedules were created by hand. As a result, the definitions of some of the most frequently used terms in construction project management, including “critical path,” have evolved. Consequently, today’s project managers must develop a more sophisticated view of what the critical path of a project schedule is and how it is most effectively used as a management tool. Project managers and schedulers must understand how scheduling has changed and how this affects their management of time on the construction project.
The general understanding of the term “critical path” is that it is in reference to controlling project time. Let’s look at some of the published definitions of “critical path.”
The Associated General Contractors of America (AGC) defines the critical path as:
. . . the longest path through the network that establishes the minimum overall project duration. The critical path is composed of a continuous chain of activities through the network schedule with zero total float. All activities on the critical path must start and finish at their planned times or the project will not complete on the calculated early/late finish of the last activity in the network.
The AGC’s explanation of the critical path goes on to state:
Failure of a critical path activity to start or finish at the planned early and late finish times will result in the overall project duration being extended. For the classical schedule calculation, it is both necessary and sufficient for an activity to be on the critical path if the activity’s total float is zero.
The AGC’s statement that, “For the classic schedule calculation, it is both necessary and sufficient for an activity to be on the critical path if the activity’s total float is zero,” is true, but it is important to understand the “classical” or “traditional” era of critical path method (CPM) scheduling. This was the era before the advent, or at least the widespread use, of constraints, multiple calendars, and other features now available in scheduling software. Because these features are now used frequently in project scheduling, the “classic schedule calculation” definition that relies on zero total float (float) no longer reliably defines the critical path of many project schedules. Because multiple calendars and constraints are commonly used in project scheduling, it is not uncommon to have activities on or off the critical path of the project that have zero float, positive float, or even negative float. The float-based portion of the AGC’s definition also implies that any activity with a float value of zero should be on the critical path. Many software programs actually allow the scheduler to assign a “zero float constraint” to any activity in the schedule. This constraint has the effect of making the float value of an activity equal zero. By using this type of constraint and the float-based portion of the AGC’s definition, any activity in the schedule could be put on the critical path. Therefore, although the float-based portion of the AGC’s definition of the critical path may be true for some projects, it is not necessarily true for all projects.
Let’s look at a few other definitions of the critical path. The following definition of the critical path is from Oracle’s Primavera 6.0 (P6):
The critical path is a series of activities that determines a project’s completion time. The duration of the activities on the critical path controls the duration of the entire project; a delay to any of these activities will delay the finish date of the entire project.
Oracle’s Primavera SureTrak 3.0a, another version of CPM scheduling software, defines the critical path as:
. . . the longest path of activities through your project. Each activity on the critical path has zero or negative total float. The duration of the critical path controls the duration of the entire project; any delay to any activity on the critical path delays the completion of the project.
Even if a person did not have these software packages, entering “critical path” into Wikipedia will lead to “critical path method,” with the following definition:
In project management, a critical path is the sequence of project network activities which add up to the longest overall duration. This determines the shortest time possible to complete the project. Any delay of an activity on the critical path directly impacts the planned project completion date (i.e., there is no float on the critical path).
There are many other variations of the definition of the critical path, in addition to the four definitions that are identified in this article. Putting these definitions aside, let’s review two common elements that often appear in definitions of the critical path. The two common elements are:
- The critical path of the project controls when the project will complete.
- If you delay an activity on the critical path, you will delay the project-completion date.
The first element is true. The critical path of the project always controls when the project is planned to finish because it is a measurement of the longest path through the schedule and, therefore, defines the earliest date when the project can finish. The same could be said for measuring the critical path through a milestone completion date (i.e., the critical path to any milestone on the project controls when the milestone is planned to be completed.) However, the second common element in the definition of the critical path is only true sometimes. The objective of this article is how to identify when the second element is not true, and then to provide an example where the path that must be progressed, to prevent delays to the project completion date, is actually (at least initially) a non-critical path.
For several years, scheduling software has allowed schedulers and project managers to place activities under separate work calendars. This is also referred to as using “multiple calendars” to schedule a project. Multiple calendars become extremely useful options for project managers when subcontractors on the job are planning to work different schedules (five-day workweek vs. six-day workweek) or if the project is susceptible to other work constraints like environmental shutdowns (no work in riverbeds during certain times of the year, wildlife constraints, winter shutdowns, etc.). Multiple calendars may affect the float value of activities in the schedule and, therefore, activities on the critical path may have varying float values. That is why any definition of the critical path should not contain a reference to the “zero float path” or the “path with the least or most negative float.” These kinds of definitions were more applicable in the “classical” era of CPM scheduling, before the inclusion of constraints, multiple calendars, and other features of modern scheduling.
Let’s look at a scheduling example that contains activities using two different calendars. Path A contains Activity 1, which has a duration of 8 months and is planned to progress based on Calendar 1. Activity 1 has a finish-to-start logic tie to Activity 2, which has a duration of 3 months and is planned to progress based on Calendar 2. Path B contains Activity 3, which has a duration of 6 months and is planned to progress based on Calendar 2. Activity 3 has a finish-to-start logic tie with Activity 4, which has a duration of 4 months and is planned to progress based on Calendar 1. Figure 1 is a graphical layout that is similar to a bar chart.
The vertical lines on the graphic identify the end of the corresponding month shown above the line. The shaded region between the end of Month 6 and the end of Month 9 identifies when Activities 2 and 3 cannot progress because of the environmental-shutdown period. Based on the work calendars for Activities 1 and 2, there is a 1-month period (all of Month 9) where Activity 1 is expected to be finished but Activity 2 cannot start because of the environmental shutdown. If Activity 2 was not susceptible to the environmental shutdown, Path A would be expected to finish 1 month earlier at the end of Month 12.
Now, let’s determine the critical path of the project shown in Figure 1. Let’s start with the definition that the critical path is that path of activities that determines when the project will complete. Path A is expected to finish at the end of Month 12. Path B is expected to finish at the end of Month 10. Therefore, the expected time to complete Path A is two months longer than Path B, and Path A is the critical path of the project. Based on our previous discussion of the two common elements of the critical path, along with the known projected durations, logic, and work calendars of each activity, Element 1 has been validated because “the critical path controls when the project will complete.”
Element 2 of many of the definitions of the critical path states, “If you delay the critical path, you will delay the project completion date.” Is this true, based on the information in Figure 1? No! What happens if the contractor does not progress Path A on Day 1? Activity 1 will be delayed 1 workday, but because of the non-work time between the end of Activity 1 and the start of Activity 2, the project is still expected to finish at the end of Month 11. Therefore, because of the multiple calendars in this example, it should be apparent that it is possible to delay the critical path of the project without delaying the scheduled project-completion date.
Now, let’s say that Path B also did not progress on Day 1. Does this have any effect on the scheduled project-completion date? Activity 3 was expected to finish on the last day of Month 6, but because it did not start, it is now expected to finish at the beginning of Month 7, right? No! Remember, Activity 3 is susceptible to an environmental shutdown between Months 7 through 9, and as a result, Activity 3 would now be expected to finish at the beginning of Month 10. How does the delay to Activity 3 affect Activity 4? Because of the finish to-start logic between Activities 3 and 4, Activity 4 would not be expected to start until Activity 3 finished at the beginning of Month 10. Therefore, Activity 4 would not be expected to complete until the beginning of Month 14. This is shown graphically in Figure 2.
So, if the project manager could only allocate resources to either Path A or B, which path should the project manager chose to not delay the project? Clearly, Path B, the non-critical path (based on durations) at the beginning of Day 1, would be the more logical path to progress to not delay the project. If Path B is delayed on Day 1, its delay to the project will be greater than if Path A were delayed on Day 1. Therefore, it appears that Element 2 of the definition of the critical path would apply to Path B, and not to Path A. However, Path B is not the critical path based on Element 1 of the definition of the critical path. Before any work has occurred on the project in Figure 1, Path A is controlling when the project is scheduled to complete; Path B is not controlling when the project is scheduled to complete. The critical path based on durations, Path A, could be delayed without affecting the scheduled project-completion date, while any delay to Path B would delay the scheduled project-completion date by at least 2 months. In fact, if Activity 3 on Path B had progressed as expected, Activity 1 on Path A could have been delayed an entire month before there would have been a delay to the scheduled project-completion date. Therefore, the statement “If you delay an activity on the critical path, you will delay the project completion date,” (as stated in Element 2) is not absolute. It also shows an example of how controlling the project-completion date and avoiding critical-path delay may be dependent on the progression of a non-critical path instead of the progression of a critical path.
This example demonstrates four important points:
- The longest path of activities through a CPM schedule determines the date when the project is expected to finish.
- The longest path in this example can be delayed without delaying the scheduled completion date of the project (at least some of the activities on the path can be delayed).
- Path B, what we would “classically” consider as the non-critical path, is not the longest path on the project. However, if the first activity on this path is delayed, the scheduled project-completion date will be delayed. In that context, some may say this path is also the critical path. Remember, once Path B experiences a delay on Activity 3, Activity 4 is affected by that delay, and Path B does become the longest path. Therefore, whether you consider Path B the critical path because it cannot experience a delay without delaying the project or you observe that once a delay occurs on Path B, it immediately becomes the longest path, the argument is academic. If Path B is delayed, the project completion date is delayed, whether or not you consider Path B to be the critical path before any progress, or lack thereof, occurs on the project.
- The project manager/scheduler’s understanding of CPM scheduling must not be based solely on definitions. They must think! They must understand how CPM scheduling works to effectively manage the project.
Conclusion
Project managers and schedulers must comprehend what the critical path truly means, and also recognize how it may or may not affect the project-completion date. Regardless of how one defines the critical path, the important aspect is that the CPM scheduling tool is designed to assist the project participants in completing the project in the most efficient manner possible, thereby preventing delays to the end date of the project. To do this requires a thorough understanding of the schedule to make sound management decisions. In addition, the critical path is dynamic in nature and may change, or shift, frequently. Critical-path shifts depend on how a project is scheduled and how the project is or is not progressing. Critical-path shifts can and very often do occur while the project is ongoing. Therefore, the project manager and scheduler need to monitor the other paths of work to determine how and when they may become critical and, as a result, affect the scheduled project-completion date. As demonstrated in Figures 1 and 2, the potential delays may become actual delays sooner than anticipated, and they may have nothing to do with the existing project critical path prior to the delay.
Published in Construct!, Volume 19, Number 1, Spring 2010. © 2010 by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.