Measures drive actions that move you toward the goal. In The Haystack Syndrome, Dr. Goldratt notes [4]:
The first thing that must be clearly defined is the overall purpose of the organization—or, as I prefer to call it, the organization’s goal. The sec- ond thing is measurements. Not just any measurements, but measure- ments that will enable us to judge the impact of a local decision on the global goal.
Figure 4.7 illustrates the cybernetic view of measures used by Dr.
Joseph Juran. The sensor makes the measure in block 2. An umpire
1 Process
2 Sensor
3 Goal
4 Umpire 4
Actuator
Output
Action
Decision
Data Requirement
Figure 4.7 Dr. Joseph Juran depicts measurement as part of a control system.
(block 4) compares the output of the process as reported by the sensor to the goal for the process. The umpire makes a decision to cause an action, modifying the process to change output and minimize the gap. That is how all control systems work and is the intent of project measurement systems, in which the goal includes the technical requirements, cost, and schedule for the project.
In The Haystack Syndrome, Dr. Goldratt defines data as “every string of characters that describes something, anything, about our reality” [4]. He defines information as “the answer to the question asked.” Dr. Goldratt suggests that the information system should incorporate the decision.
The improved measurement system for CCPM follows the practice established by Dr. Goldratt for production operations. It uses buffers (i.e., time) to measure task chain performance. Recall that the end of the proj- ect buffer is a fixed date: the project delivery date. For buffer management purposes, you also fix the ends of the feeding buffers. You determine project buffer penetration by asking people working on tasks, “When will you be done?” That allows you to project forward using the downstream task duration estimates to predict how much of the buffer would be used up if they complete at that time.
You size the buffers based on the length of the task chain they project.
Buffer sizing uses the uncertainty in the duration of the critical chain tasks to size the project buffer. Likewise, uncertainty in the duration of the feeding chain tasks determines the size of each critical chain feeding buffer (CCFB). CCPM sets explicit action levels for decisions. The decision levels are in terms of the buffer size, measured in days:
1. Within the first third of the buffer: Take no action.
2. Penetrate the middle third of the buffer: Assess the problem and plan for action.
3. Penetrate the final third: Initiate action.
Those measures apply to both the project buffer and the CCFBs.
Figure 4.8 shows an example of using the buffers. The three Xs show three potential amounts of buffer penetration corresponding to the above criteria.
Project teams monitor the project buffer and each CCFB at the appro- priate time intervals for the project, usually weekly but at least monthly.
You can update the buffers in a relatively short weekly meeting or use E-mail status reports. For this tool to be fully useful, the buffer monitoring time must be at least as frequent as one-third the total buffer time. If the buffers are negative (i.e., the latest task on the chain is early relative to schedule date) or less than one-third the total buffer late (e.g., less than 10 days if the total buffer is 30 days), you do not need to take action. If extended durations penetrate the buffer between one-third and two- thirds, the project team should plan actions for that chain to accelerate the current or future tasks and recover the buffer. If the task performance penetrates the buffer by more than two-thirds the buffer size, the project team should take the planned action. Through this mechanism, buffer management provides a unique anticipatory project management tool with clear decision criteria.
Project managers update the buffers as often as they need to simply by asking each of the performing tasks how many days they estimate to the completion of their task. They do that without pressure or comment on the estimate. They expect the estimates to vary from day to day and some of the tasks to exceed the original duration estimates. As long as the resources are working on the tasks with the CCPM task performance paradigm, managers evaluate them positively, regardless of the actual duration.
An enhancement in the use of the buffer for long critical chains is to plot trends for buffer utilization, as shown in Figure 4.9. The buffer measure then becomes functionally similar to a control chart and can use similar rules. That is, any penetration of the red zone requires action.
Four points trending successively in one direction require action.
0/3 1/3 2/3 3/3
Project buffer
CCFB-1
CCFB-2
X
X
X No action
Plan
Act
Figure 4.8 Buffer penetration provides action decisions.
People have suggested that project managers also monitor buffer penetration as a percentage of critical chain use. The idea is that the man- ager should not use up the project buffer too early in the project. Trending buffer penetration has several advantages over that approach. Perhaps the most important advantage is that it is simpler to interpret. A second advantage, especially important if the project task processes are not in statistical control, is that trending preserves the time history of the data.
That is important information to help improve control of the processes.
Updating the buffers requires that you maintain project status versus your plan in terms of the tasks completed. It is also a useful direct measure of project performance.