Ihab Saad – Earned Value Analysis EV
AI: Summary ©
AI: Transcript ©
Music, good afternoon,
and today we're going to talk about a new subject related to
project scheduling, which is measuring the project performance
through a technique called earned value analysis.
In a recent study by the construction industry Institute,
they found out that cost and schedule are two very
representative measures of the Project Health, and they found
that there's a direct correlation between the scheduled performance
and the cost performance as indicators of how the project is
performing in general. So the closer the cost, the actual cost
and schedule are to the planned values, the better the project
performance is and will be at the end. So basically, cost and
schedule are definitely two measures, two active and valuable
measures to gage the Project Health and its performance from
time to time on a regular basis, to try to forecast the end result
for that particular project. However, measuring cost and time
or cost and schedule performance usually have some problems, if not
not done properly. Among these problems is that trying to measure
the schedule and cost performance for the sake of reporting on
general project performance in general project managers sometimes
fall in the trap of comparing apples to oranges, and this
primarily results from comparing two moving targets, or making a
comparison between two moving targets at the same time without
having something to compare to. So for example, if I want to measure
the speed of something that's moving, I have to compare it
either to something that's stationary or to something that's
moving at a certain speed that I already know. So I can measure the
relative speed or the relative difference between the two speeds.
If we do not have such benchmarks, we will not have valid comparisons
when it comes to time and schedule. So successful project
managers at the very beginning should establish agreed upon
baselines related to cost and time, and these are going to be
basically the approved budget at the very beginning of the project
and the approved baseline schedule that in general, the general
contractor submits to the owner for review and approval. Once
these are established, once these are approved, they become our
benchmarks or our measuring lines, and we can start measuring
progress from these fixed benchmarks. We also have to
establish some thresholds to measure our variances. So for
example, we know that sometimes we might be a little bit above
budget, little bit below budget or ahead of schedule or behind
schedule. So we have to establish thresholds, and they can be color
coded, for example. So green is anywhere if we're performing at
schedule or on but on time plus or minus, let's say three to 5% so if
we are in a range, little bit ahead of schedule, little bit
behind schedule or right on schedule, we are fine. So this can
be green.
If the range of variance is greater than or anywhere between
five and seven or 8% for example, it can be yellow. We are
approaching the level of danger or warning. Anything above that is
going to become red. So again, color coding that will go will
show us, just with a visual inspection, how the project is
performing. So we have to establish these thresholds from
the very beginning and monitor the project accordingly.
So what's earned value? Earned value analysis used to also be
known as C slash, SCSC, which is which stands for cost slash,
schedule control systems, criteria, very complicated name
that was invented by the Department of Defense because they
were the first ones to use it to measure the performance on their
projects. So in In brief, we call it earned value analysis, or EV.
It builds on the concept of comparing apples to apples and
oranges to oranges. So again, when we're going to make a comparison,
it has to be a valid comparison, and not just comparing two numbers
that are totally unrelated.
It enables the project managers, through a simple and effective
analysis, to get timely updates and forecasts of projects, time
and cost performance. The key word here is timely, because if you get
the update too late, unfortunately, its value is very
much reduced because you cannot make any corrective actions or any
improvements on the project. So if you get all of these updates at
the end of the project that's past history. It's just a documentation
of the project performance, without the ability to change any
of its outcomes. So we have to do these on a regular basis and in a
very timely fashion. So if we're going to update our schedule on a
weekly basis, for example, we can generate that report or that
analysis on a weekly basis. Same.
Thing, it's very highly recommendable that whenever you
update your schedule, you also update your budget, so you can get
a comparison of time and cost at the same time on a very regular
basis.
So to discuss earned value, let's start by this short story or
example on a project that is under construction.
The president of a company received a call. The CEO of a
large company was contacted by his CFO, the Chief Financial Officer,
usually an accountant, who informed him of the following,
saying that, sir, according to the approved schedule and our accurate
cost estimate, until this morning we should have spent $38.5 million
on the project. So the cost expenditure up till this morning
should have been $38.5 million however, according to my latest
calculations, until this morning, and this is we're talking about an
accountant who's very meticulous, very accurate, until this morning,
we have actually spent $40.3 million
so to the account that it appears that we have a big variance here.
That's a considerable variance, by the way. So he says, Sir, this is
a serious matter that requires your immediate attention. I
recommend you fire the project manager and maybe also the Vice
President for Operations. There's no
love lost between the two. So what do you think if the CEO of the
company listens to that very accurate and meticulous
accountant, Chief Accountant, Chief Financial Officer, what
should they do if you were in that situation? What What would you do?
The question is, do we have enough information to make an educated
decision about something as important as firing someone or
hiring someone or changing the outcome of the project? So let's
see whether these numbers, as accurate as they are, as correct
as they are, really reflect the status of the project, or is there
something missing? And here we're going to start with what we call
the tale of
three fruits.
So basically, through this story, we have not only two fruits,
apples and oranges, but we must also look at the third fruit. So
in the previous report, the CFO made a classical mistake of
comparing an apple to an orange. He looked at two numbers, two
different values, which really represent an apple and an orange.
So there's nothing to compare between an apple and an orange.
We're going to see that in just a second. So the apple,
the apple represents the budgeted cost to be spent on the scheduled
work, the $38.3 million or whatever it was so. And this is
going to be called the budgeted cost for work scheduled. Budgeted
cost because it was based on the estimate for work schedule,
because it was based on the baseline schedule that was
approved by the client. And that number is the value that the CFO
told the owner, until this morning we should have spent. Should have
spent means it was a budgeted cost, and until this morning,
which is the work done to date, according to the schedule, based
on the scheduled performance. And then he told him, yet until this
morning, we have actually spent. So now he's talking about actual
expenditure we have actually spent. It was 40 point something
million dollars. Again, big difference. So the orange
represents the actual cost spent on the actual amount of work
performed,
or actual cost for work performed, AC, WP, so we have two four letter
words, BCWS and acwp. What do we have in common? If we try to split
these words into two parts. In the first part, we have BC, for the
second one, we have AC. They are not common, budgeted costs and
actual costs are quite different. The second half is work schedule,
Ws and work performed. WP again, work scheduled is what we should
have done. Work performed is what we have actually accomplished. So
again, they might be different, so there's nothing
comparable between the two terms. Therefore, what we're going to do
is we're gonna slice this apple into two halves, and we're gonna
slice this orange into two halves, and we're gonna look at the hybrid
fruit that we're gonna call the orpal, or the appenge. Orpal, half
orange, half apple, apench, half apple, half orange. Either term is
fine with me.
So the solution is the average, or the orphan, which looks something
like this, half apple, half orange. When we split this apple
in halves, we took the left half of the apple and the left half of
the Apple was BC budgeted cost. And then when we took the half of
the orange, we took the right half of the orange, which represents
word.
Cost, which means we are performing under budget, which is
having some savings, and that's definitely good. Scheduled
variance, or SV, is equal to the earned value minus the planned
value. Again, negative is bad, because
planned value
is greater than the earned value, which means we are behind
schedule, and positive is good, which means that the earned value,
what we have earned, is greater than the planned value, which
means we have achieved more than what we were supposed to do in the
same period of time, which means that we are ahead of schedule.
The cost performance index, sometimes also called cost
performance ratio. So ratio or index is the same thing, and it
results from the Division of two different terms. And in this case,
instead of subtracting one term from another actual cost from
earned value, we divide earned value by actual cost. So in this
case, we have something to be compared to one under perfect
conditions, the earned value is equal to the actual cost if we are
performing exactly according to budget. So in this case, comparing
to to one, if the earned value is greater than the actual cost,
which means that we have earned more than what we we actually
spent, then greater than one is good and less than one is bad,
because the actual cost is going to exceed the earned value.
Similarly, for the schedule, if the earned value is greater than
the planned value, greater than one is good, because we would be
ahead of schedule. Less than one is bad.
The estimate at completion can be calculated in three different
ways. Either it's the budget at completion divided by the cost
performance index, or it is the actual cost plus the estimate to
complete, or it is the actual cost plus budget at completion minus
earned value. Either one of these three terms can give us the
estimate at completion, the estimate to complete is equal to
estimate at completion minus the actual cost.
Variance at completion is equal to budget at completion, our original
budget minus estimate at completion based on the current
progress. Positive is good, which means that we are saving money.
Negative is bad, which means we're going to finish above budget and
to complete performance index, or TCPI, is budget at completion
minus earned value divided by budget at completion minus actual
cost. Now, from these different terms, the most commonly used
terms are the first four, which are the cost variance, scheduled
variance, cost performance index or cost performance ratio, and
scheduled performance index or scheduled performance ratio. We're
going to see in a minute how to calculate these.
So now let's look at that graph representing the
time on the horizontal axis and cost on the vertical axis. Now
when we look at this curve, this reminds us of what we used to know
as the S curve, or the progress curve, and the final point on that
progress curve is our budget at completion of BAC. Now that dotted
vertical green line represents our cutoff date, as we used to do in
updating, in Project updating, so as of this cutoff date, we have
different values based on our budget and based on our schedule.
Both plans, we should have spent $38.5 million according to that
up to that cutoff date. However,
our actual cost for work performed or actual cost is $40.3 million
and now this is the classical mistake when the CFO made the
comparison between these two moving targets. Notice that this
curve is not continued, because this is actual and we don't know
how are we going to perform in the future. We can, however, forecast
our future based on our current progress. But up to this point,
we're just reporting past history of actual performance, and we have
not projected that into the future yet. And now we're looking for
that third term, something in between these two, which is now
this is the forecast, as you can see, and it's a dotted line that
shows that our estimate at completion here is going to be
above the budget at completion if we keep performing the same way.
So it seems that we're not doing well. Now let's look at the third
number, the orphan that we're looking for, and we're going to
talk later about how to perform to calculate that number. But if that
number were here, bcwp, or earned value, if it was some somewhere
between these two numbers, we're going to look at different
scenarios. What if that point is above both curves? What if it is
below both curves? What if it's in between and what? And what's the
respective order of these points. How does that affect our Calc?
Relations. We're going to talk about that in a minute, but let's
assume for the moment that the budgeted cost forward performed
was calculated equal to 39.8
so now let's look at according to the equations that we discussed on
the previous slide. Let's calculate our variances and our
indices. So the first variance is the comparison of the apple to the
orange, which is equal to the accounting variance. What we call
accounting variance. The reason I did not include it on the previous
slide is that we as project managers and cost engineers and so
on, we should not use that number. This number is sometimes
mistakenly used by accountants, but we as project managers should
not use that number. We should know how it's calculated, but we
should not use it. The accounting variance is equal to actual cost
for rule perform, minus budget, cost for schedule, and here it
shows that we are far above our budget. But again, that's not the
measure that we're looking at, because that does not measure time
in any way.
Now let's start looking at comparable numbers. Here we have
bcwp, which is the earned value, and BCWS, which is the PV or the
planned value. What do they have in common? BC and BC, if we remove
the common half of each one of these words, we end up with word
performed compared to work scheduled.
So if the word performed is greater than the work scheduled,
if we have a positive number from this equation, word performed is
greater than the work schedule, that means that we are ahead of
schedule, ahead of schedule by an amount of work equal to the value
or the difference between these two, which is about $1.3 million
so our scheduled performance is fine. We have achieved more work
than we were supposed to do or we were planned to, up till that
cutoff date. So schedule wise, we're doing well. So this is the
scheduled variance, and then we're going to compare the cost
variance. Again, comparing an or an orpal to an orange, the common
part is the or, which is, in this case, the WP and the WP. So
comparing earned value to actual cost. We're going to remove the
common term WP. So we're comparing BC to AC, budgeted cost to actual
cost. In this case, the actual cost is above the budgeted cost,
which means we're going to get a negative value here. And as we
learned from the previous slide, negative value means bad we are we
are over budget. So in this case, we are over budget by about half a
million dollars, which is difference between 40.3 and 39.8
so basically, if we want to report on the actual status of this
project, we can use just two words to report a general very brief
report. Is it good, good, good, bad, bad, good or bad, bad. So is
the schedule good and the cost good? So it would be good, good,
or the schedule is good and the cost is bad, which is good, bad,
exactly what we have here. Or is it bad, good, the schedule is bad,
but the cost is good or bad, bad, which would be if that bcwp is the
lowest point is below acwp And BCWS. So if the earned value falls
below both other point, the two other points, if it's the lowest
point, then that would be a bad bet. However, if that point is
above both other points. If above the two points, then it would be
good, good, because in this case, we would be ahead of schedule and
below budget, which is which would be a perfect situation.
So the analysis results from the previous graph, we can conclude
the following, the scheduled variance is $1.3 million positive,
so we are ahead of schedule by an amount of work worth $1.3 million
which is good, and the cost variance, we have a negative
number, negative half a million, which means the project is above
budget by point 5 million, which is bad. Therefore the final
situation for this project is good, bad
or bad, good, depending on if you compare the cost first and then
the scheduled second.
The variances, however, however important as they are, do not show
the full picture, because what if, for example, I get a cost variance
of negative half a million dollars. Okay, we are above
budget. But how drastic the situation is, how bad the
situation is. If I were to tell you, for example, that the value
of the project is $1 billion
so half a million compared to 1 billion,
that's negligible, we can recover that covid.
In our schedule is going to be represented by a cost item, cost
center or cost unit. So whenever we report, we're going to report
on time and cost at the same time.
In case of emergency, use the Pareto law 8020
which basically says that 20% of the activities in the project
represent 80% of the cost of the project. So I, if I cannot perform
time and cost control over 100% of the activities, at least, I'm
gonna focus on the 20% of the activities that represent 80% of
the total project cost. And that would give me an acceptable level
of control. It's not perfect, but 80% is not too bad, either,
especially that it's achieved with only monitoring, monitoring 20% of
the activities. And how do we get these 20% of the activities? This
is for another discussion.
Now here, here's an example that shows a very simple example on a
network represented by six activities, A through F. And this
is the duration in weeks for each activity, and this is the
precedence, which means what are the IPAs? So A and B are the
beginning activities for this project. And then we have here the
succession of the different activities, and we also have
budgeted cost per week. And this is obtained from our estimate.
Budgeted cost per week. How much do we expect to spend on this
activity per week according to the estimate? And then here we have a
blank column where we have to calculate the total budget. And
very easily, the total budget is going to be obtained by
multiplying the budgeted cost per week times the duration in weeks.
So basically, for activity A, for example, it would be $690
for activity B, it would be six times 280, and so on and so forth.
By adding these numbers together, what we're going to get is, what
if you said BC WP, if you said BCWS, that would be half correct
and half incorrect, because it would be BCWS only at the end of
the project, not at the time we are measuring BCWS at the end of
the project is basically what we call BAC, or the budget at
completion. Therefore the sum of these numbers is not going to be
used in our equation whatsoever.
Now the problem is, or the question is, at the end of week
nine, which is our cutoff date, our green line, the actual project
status was as follows. These numbers were calculated from the
site, and this progress was reported from the site for
activity, it was 100% complete, so PCT, 100%
therefore it was complete. B, 100% C, 75%
D, 0% E, 60% and F, 50%
and these are the actual costs of each one of these activities,
actual cost, actual cost up to the end of week nine. So actual cost
for the work performed up till the end of week nine. Therefore, if we
add these numbers together, that's going to give us the acwp at the
end of week nine.
Now we need to calculate the earned value. What is the earned
value up till the end of week nine as well. So for each activity
here, we have assumed a linear relationship between cost and
schedule, and that the BCWS is determined from the earliest start
date. So assuming a linear relationship between cost and
schedule, which means that the cost is uniformly distributed
along the duration of the activity. That's why we mentioned
that the cost is the same for each week. This is the assumption that
we need. So the question now is, how are we going to determine
whether this project is ahead of schedule, behind schedule, on
budget, above budget, below budget, and we're going to try to
measure the cost variance, scheduled variance,
and the CPI and the SPI. So what we need to do here is to draw a
cumulative cost curve for this project, showing the duration and
cost, which is basically our s or progress curve. And then we need
on that curve to identify acwp or actual cost BCWS, or planned
value. Bcwp or earned value, CV, cost variance, SV, scheduled
variance on the curve. And we need also to determine the values of
the CPR or CPI cost performance ratio, or cost performance index,
and SPR or SPI schedule performance ratio or scheduled
performance index. And finally, in your expert opinion, what is the
status of this project, schedule wise and cost wise. So are we
ahead of schedule? Behind schedule, above budget, below
budget? Is it good, good, good, bad, bad, bad, good or bad, bad.
Ed, this is what we're going to answer through solving that
example in the next presentation.