davegb said:
I'm wondering, is this useful in terms of knowing your project's
status? Won't this practice undermine the purpose of doing EVA? On the
one hand, if there's one task under a summary line what is way behind
on cost or time, and 3 others that are a little ahead, they could zero
each other out in the combined EV at the Summary level. You wouldn't
know you have a problem. I've always thought that one of the reasons
for doing EV is that when you had a -CV or -SV, you could drill down to
individual task CV or SV and see which tasks were causing the problem.
And if there are no offsetting ahead-of-schedule tasks, if you average
them down from a Summary line, you can't tell which of them is the
source of your problem. It would be very frustrating to know you have a
problem, but not be able to figure out what task(s) are the cause.
Hope this helps in your world.
Dave,
I certainly don't know Jim's impetus but from my working experience,
yes, earned value at summary level can be a smart way to manage. Here's
why.
In some industries, (perhaps the majority), contracts with earned value
clauses are proposed, negotiated and planned at a WBS level that is
often only a few levels deep. On the other hand, any good project plan
will take the actual working detail much lower. The customer and upper
management are more interested in the "big swingers" than in the detail
"bumps". That is not to say that the cost account manager (CAM) who is
responsible for the detail, need not worry about individual tasks that
are in cost or schedule trouble. On the contrary, the CAM's job is to
manage the work and know the impact on the program's overall end goal.
At the reviewing level (i.e. customer/management), it isn't all that
important if CAMs have ahead of cost/schedule tasks that may be
canceling out behind cot/schedule tasks. In my experience at least, if a
CAM is "even" from an earned value standpoint, chances are he has a good
handle on his part of the program. If a CAM is NOT managing his tasks,
the probability of his earned value balancing out month after month (or
whenever the review cycle occurs) is extremely low.
For reference, every project I worked on was managed like this. It kept
the responsibility at the detail level and minimized the tendency to
have a micro-managed project. Of course a very important part of the
process is the periodic review (ours were monthly - tracked with our
statusing period).
That's my 2 cents.
John