Earned Value Formulas for Monitoring Project Progress


There are three key earned value components that have to be mastered before the earned value methodology can be understood and used. They are the planned value (PV), the actual cost (AC), and the earned value (EV). At this stage, it is easiest to think in terms of tasks than in terms of the whole project. How these costs roll up to the project cost will become evident as we progress in the discussion.

PV is the amount that you estimate the task will cost. It is the amount that you estimate the cost will be for a task's planned schedule, or to say that another way, it is the cost of the scheduled work.

AC is the actual amount of money paid for the work as it progresses. This information comes from invoices paid, labor costs, that is, time sheet data, and materials bought for the task at hand. AC is the actual cost of the work performed on the task.

EV is the amount of money budgeted for the work that is actually performed. This term is the most difficult one to conceptualize because it presents a new way of thinking about progress. The simplest way of think about EV is this: Suppose you have a task that you estimate will cost $1,000 and will take twenty weeks to complete. To simplify the explanation, we will break the task into one-week elements, as shown in Exhibit 9-1.

Exhibit 9-1: Determining PV and EV for task work progress.

start example

click to expand

end example

In the example shown in Exhibit 9-1, I have broken the task into weekly segments to better explain the concept of PV and EV, but especially EV. The total planned value cost of the task is, of course, $1,000, but before the task is completed, the "to date" PV will be determined by adding the costs of the various segments to the assessment point. In this example, we are going to assess the progress at the end of the fifth week. Hence, the PV to date is five times fifty dollars, or $250. If the project were on time, that is, on the planned schedule, then the EV would also be $250. However, in the example we are saying that we are actually ahead of schedule and have completed the equivalent of seven weeks of work. So the EV is the sum of the cost segments actually completed to date, or $350. Another way of calculating EV is to multiply the percentage of workweeks completed times the total task PV (7/20 $1,000 = $350).

With this understanding of the three primary components (PV, AC, and EV) of the earned value management methodology, we can go on to an IT example that shows how to actually monitor the project's progress using earned value.

Monitoring Project Progress with Earned Value Analysis

Earned value analysis is, simply stated, determining how well the project is doing in comparison with the original plan. The use of an example best illustrates earned value analysis. Exhibit 9-2 is an example of an IT project consisting of several tasks. Naturally, this is only a small representation of a project, but it should suffice to demonstrate the principles of the technique.

Exhibit 9-2: Earned value analysis of an IT project.

start example

click to expand

end example

In the example, we have a small IT project that is fourteen weeks in duration. At the six-week point, we are going to make an assessment of the project's progress. (Note that the project manager and his team will have made similar assessments at the end of each of the previous weeks, but to demonstrate the analysis technique, we are only going to look at the six-week assessment.)

The most convenient way to collect earned value data is in tabular form. Exhibit 9-3 is an example of common earned value tables of the type provided by most project management software programs.

Exhibit 9-3: Earned value analysis table.

start example

Earned Value Analysis

Task Name

Col. A

% Complete

Col. B

Total Task PV

Col. C

To Date PV

AC

EV

(Col. A Col. B)

CV

(EV - AC)

SV

(EV - PV)

Requirements Definition

100

$500

$500

$600

$500

- $100

0

Systems Architecture

100

$5,000

$5,000

$4,500

$5,000

$500

0

Software Design

67

$3,000

$2,010

$2,100

$2,010

- $90

0

Hardware Design

83

$2,000

$1,340

$1,600

$1,660

$60

$320

Software Coding

25

$10,000

0

$3,000

$2,500

- $500

$2,500

Totals

$20,500

$8,850

$11,800

$11,670

- $130

$2,820

end example

The point of the analysis is to determine whether we are on schedule and on budget. From Exhibit 9-2, it is easy to see that we are actually ahead of schedule. We have completed the work that was planned for completion at the sixth week, we have begun the software coding task, and we have completed more of the hardware design task than was anticipated. However, there is no way to look at this modified Gantt chart and determine whether we are on budget—that has to be done by comparing what is completed against what we have spent. But when we look at the collected data in the earned value table (Exhibit 9-3), the column labeled CV (cost variance) tells the story—we are over budget.

Because the CV is a negative number, we know that the project is over schedule. Remember that a positive CV indicates an under budget condition, negative indicates over budget, and zero means the project is on budget. Likewise, a positive SV (schedule variance) means the project is ahead of schedule, which is the case in our example. A negative SV would mean the project is behind schedule, and zero means the project is on schedule.

Notice that in the analysis, we have to determine where in the project's development we planned to be at the time of the assessment. We also should know how much work we actually have accomplished to date. The first two tasks, requirements definition and systems architecture design, are completed when we take the assessment at the sixth week. So, at this time, we had planned to spend the money budgeted for the tasks, and we completed the work. So both the PV and the EV for each of the tasks is equal to the amounts estimated, $500 and $5,000, respectively. The next task, software design, is a three-week task, and we planned to have completed two weeks of work at the assessment point. The shaded area indicates that we did, in fact, complete the two weeks, or 67 percent of the work. So we can determine the PV to date and the EV to date by multiplying 67 percent by $3,000, the estimated PV for the task.

The next task, software coding, is a little different—we had not planned to begin that task until the sixth week, or the time of the assessment. So the PV for the task is zero, but the shaded area indicates that we actually managed to begin the work sooner and have finished one week (or 25 percent of the coding) already. Hence, the PV is zero, but the EV is 25 percent x $10,000, or $2,500.

The hardware design task also has gone quite well, and we can see from Exhibit 9-2 that we have completed a half week more work than planned. So the PV to date is 67 percent (2 weeks planned progress divided by 3 weeks total task duration), and the EV is 83 percent (2.5 weeks actual progress divided by 3 weeks total task duration). The other tasks were not planned to begin, and no work has been accomplished on them.

Notice on the earned value table that the totals of PV, EV, and AC can be used to determine the total cost variance and schedule variance for the project as well. Not only does this serve as a check of your math, it also is more meaningful for senior management. After all, their interest is in the project's progress; your interest is in the tasks that are impacting the project's progress.

Other important measures in earned value are the cost performance index (CPI) and schedule performance index (SPI). Because these measures represent the project progress in terms of a percentage rather than a dollar figure, they are important and have more meaning to senior management or to financial people.

CPI is determined by the following formula:

EV/AC = CPI

So in our example, the CPI is:

11,670/11,800 = .99

This number can be interpreted to mean that for every dollar we spend on the project, we are getting ninety-nine cents back. If the CPI ratio is less than 1.0, then it indicates that the project is over budget, which is the result we got when we calculated cost variance. But the difference here is that the CPI of ninety-nine cents provides a better relative sense of how bad the situation is. In other words, comparing a cost variance equal to -$130 with a CPI of ninety-nine cents, we know that, although we are over budget, we are actually very close to being on budget. We don't have to make any major changes to the way we are managing the project.

The other measure, SPI, gives a similar indication of how the project is progressing against the planned schedule. The SPI is represented by the formula:

EV/PV = SPI

So from our example again, the SPI is:

11,670/8,850 = 1.32

A number less than 1.0 would indicate the project is behind schedule, equal to 1.0 means the project is on schedule, and a number greater than 1.0 means we are ahead of schedule. In this case, we are about 30 percent ahead of where we had planned to be at the sixth week.

These measures, CV, SV, CPI, and SPI are excellent ways to monitor the project's progress. These measures, along with others, are equally good for controlling the project; that is, they can be used to keep the project within acceptable bounds.

Project Control Through Variance Management

The key to controlling a project is using variance management as determined by earned value analysis. One way to use variance management is to determine the acceptable bounds away from the planned budget and schedule that we can tolerate. Generally, about 10 percent variation above or below the planned budget or schedule is about the maximum that can be tolerated. From experience, we know that once a project gets more than 15 percent off its plan, it can not be recovered to more than about 10 percent of the original plan. So the key is to keep the project as close to its original plan as possible, and certainly close enough to recover it, if it does get off track.

Once the bounds are set for the project's budget and schedule, then it is a simple matter to determine how close each task is to its plan whenever an assessment is made. The CPI and SPI measures are perfect for tracking how well the project is doing. Suppose you have a project that has the CPI and SPI values shown in Exhibit 9-4.

Exhibit 9-4: CPI and SPI table for controlling the project.

start example

Assessment Periods

PV

AC

EV

CPI

SPI

Week 1

$10,000

$10,300

$9,500

.92

.95

Week 2

$6,000

$5,950

$6,250

1.05

1.04

Week 3

$7,500

$7,200

$7,400

1.03

.99

Week 4

$4,500

$4,650

$4,500

.97

1.0

Week 5

$9,250

$9,300

$8,950

.96

.97

end example

Once the CPI and SPI are calculated for each period, it is an easy task to plot them on a chart like the one in Exhibit 9-5. This kind of plotting is easily done in any kind of spreadsheet. Notice that a 10 percent upper and lower limit line has been placed on the chart. Both CPI and SPI can be plotted relative to 1.0. Remember that 1.0 means on budget and on schedule, so any variation from that point is a potential for concern. However, as long as CPI and/or SPI remain within the 10 percent limits, the project is considered to be progressing according to plan, or at least within acceptable variations from the plan. In our example, the CPI and SPI are close enough to the 1.0 line that there is no need for concern. If it gets off course, it can be quickly brought back onto the planned track. So the example shows a pretty stable and well-run project. However, if all the measurements (either CPI or SPI) are all on the same side of the median line, and if they are slowly getting further away from that line and closer to one of the limits, then the project is clearly going out of control. In that case, there is a need for immediate action to bring the project back on track.

Exhibit 9-5: Tracking CPI and SPI.

start example

click to expand

end example

Many inexperienced project managers often question whether it is necessary to have a lower limit, especially for CPI. After all, isn't it a good thing to have a project under budget? Well, maybe. The problem is that being under budget can mean that a task, or a part of a task, is not done or not completed. So any time a project is under budget, it is incumbent upon the project manager to determine the cause. It is just as important to determine why the project is under budget and ahead of schedule as it is to determine why it might be over budget and behind schedule.

Controlling the project also means understanding how the present project status impacts the future project status. This information is determined by two earned value calculations—the new estimate at project's completion and the estimate of money needed to complete the project.

Estimate at completion (EAC) is one number that not only the project team is concerned about—it is also very important to senior management. Consequently, this is one of the numbers that is always requested in status reports.

The EAC is calculated by the formula:

BAC/CPI = EAC

In this case, BAC is the budget at completion, or the total budget of the project. Returning to the example of Exhibit 9-3, the BAC is $20,500, or the sum of the estimated cost/budget of each of the tasks. The CPI for that example is determined by the formula:

EV/AC = CPI

11,670/11,80 = .99

Hence, EAC is equal to:

20,500/.99 = $20,707

In this case, the new budget at completion, $20,707, is greater than originally estimated but not by so much that there is a need for concern. In fact, considering that the project is nearly halfway completed, a new estimate of only a little over $200 more than the original estimate is extraordinary—in short, the original estimate was a good one.

Another calculation that is also very important to senior management, and of particular interest to the comptroller, or financial officer, is the estimate to complete (ETC). This number tells the senior management how much money is needed to complete the project. It is important because it is the basis for determining how much cash flow is required until the project is completed.

ETC is calculated with the following formula:

BAC (or latest estimate) - AC = ETC

In other words, we simply take our latest estimate of the project's cost/budget and subtract the amount of money that has actually been spent to date. In our example, the new EAC is $20,707. The amount of money actually spent to date is found in the table of Exhibit 9-3 and is $11,800. So the ETC is found by the formula:

ETC = 20,707 - 11,800 = $8,907

So to finish the project, we will need $8,907.

All this information is needed to control a project. The controlling process includes not only determining the earned value data, but also reporting it in a way that facilitates controlling it. Progress, or status, reporting provides this important link between analysis and managing the project to maintain the planned budget and schedule.




Managing Information Technology Projects
Managing Information Technology Projects: Applying Project Management Strategies to Software, Hardware, and Integration Initiatives
ISBN: 0814408117
EAN: 2147483647
Year: 2003
Pages: 129
Authors: James Taylor

Similar book on Amazon

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net