Scope Creep / Earned Value Question for Steve (or anyone)

[Joe]

I am managing a software development project. I also
provide Earned Value such as PV, EV, AC (BCWS, BCWP,
ACWP), SPI, and CPI data to upper management. Often
times, there are additional task that need to be added to
a project plan (scope creep) during the project life
cycle. In terms of baselining and earned value data, what
is the best way to track the project plan? For example,
should I baseline new task items as they are added to the
plan after the initial baseline? (Note: I wouldn't re-
baseline the entire project, just the new task items that
were added). Or should I not re-baseline every time a new
task is added. What I want to know is what effect does
either scenario have on the Earned Value data
(specifically SPI and CPI). My goal is to show management
the impact scope creep has to a project plan. If the SPI
and CPI are more impacted by not re-baselining, would want
to do that to show managers the impact.

 

[Jack D.]

I am managing a software development project. I also
provide Earned Value such as PV, EV, AC (BCWS, BCWP,
ACWP), SPI, and CPI data to upper management. Often
times, there are additional task that need to be added to
a project plan (scope creep) during the project life
cycle. In terms of baselining and earned value data, what
is the best way to track the project plan? For example,
should I baseline new task items as they are added to the
plan after the initial baseline? (Note: I wouldn't re-
baseline the entire project, just the new task items that
were added). Or should I not re-baseline every time a new
task is added. What I want to know is what effect does
either scenario have on the Earned Value data
(specifically SPI and CPI). My goal is to show management
the impact scope creep has to a project plan. If the SPI
and CPI are more impacted by not re-baselining, would want
to do that to show managers the impact.

The value of saving a baseline is that it is a baseline against which change
can be measured. If you change the baseline you have changed what you are
measuring against. Certainly there are occasions where it is reasonable to
change a baseline (a large and approved change of scope which renders the
old baseline meaningless) but generally it should not be changed.

The best way to understand how Earned value in Project works is to create a
small test project and experiment with it. It will show you the effects much
more clearly I can describe them. Using a test project you can answer your
own questions. It really is invaluable to experiment.


--
Please try to keep replies in this group. I do check e-mail, but only
infrequently. For Macros and other things check http://masamiki.com/project

-Jack Dahlgren, Project MVP
email: J -at- eM Vee Pee S dot COM


+++++++++++++++++++

 

[Steve House]

Which version of Project are you using? 2002 or 2003 gives you multiple
baselines to handle just the situation you describe. I'm of the opinion,
and others may differ, that the baseline essentially represents the project
you intend to do. If tasks run long or short, start late or early, or
require resource substitutions that create a cost variance you compare to
the original baseline to keep track of those variances etc. But a project
is fundamentally defined by its scope and if the scope changes due to scope
creep, the project itself that you are intending to do has changed. First
of all I'd keep the original baseline for comparison purposes so using a
version that allows multiple baseline would be a very good idea. After
adding the new task(s) to accomodate the new deliverables, I'd re-baseline
for all tasks after the new task's start but retain the original baseline
for all tasks before (that way any variances that have occured before the
scope change are still retained after. Now you can go to the boss with good
numbers like "Look here - we were running a SPI of .98 until Joe insisted we
paint the database pink but now because of the additional work we're at
....."

 

[Derrick Robinson]

your SPI and CPI (cost and schedule variances) are as good as the degree in
which your estimates of percentage completion reflects reality. What ever
ratio you get for both can sometimes result in one favorable and one the
other not. You can be ahead of schedule and behind in cost, or vice versa.
You can combine the two indexes to make a form of critical ratio or
cost-schedule index. CSI = (CPI) (SPI). SPI < 1 indicates a problem.
In addition you can use the CPI to estimate the cost to complete the work.
That will give you the extra cost cause by scope creep.
Then you have your impact.
ECI (estimated cost to complete) = (BAC - BCWP)/CPI
BAC=budget at completion.
DR

 

[Sarah]

Joe,

I believe Earned Value analysis is based on a comparison of the actual
(or scheduled) to the baseline. So if you do NOT baseline new tasks,
they won't have any effect on your EV analysis.

Sarah
(sarah_kiko@(removethis)cinfin.com