I have found that baseline data gets corrupted when the tasks are moved in
schedules that have been baselined.

To reproduce the problem, do this in a new project:

1. Create 1 summary task, having one sub task (1 day will do).
2. Create a resource with a cost per hour ($1 will do) and assign that
resource to the sub-task.
3. Baseline the whole project and save.
4. Set the project status date to the day before the summary task start date
(EVM calcs are done based on this date).

Now, switch to the EVM table in the Gantt chart view and add the CPI and SPI
columns. The CPI and SPI will be 0 on summary and sub-task - this is correct.

Open the bottom pane so that you have a combination view with the top being
the Gantt chart with the EVM table and the bottom being the standard task
usage view. Add the actual work row to the task usage pane. Now you should be
able to see 8 hours work and a blank for actual work, on the day for the
subtask.

Select the sub-task. Enter an amount of hours equal to the work value, into
the actual work row, on the expected date for the subtask (in the task usage
pane). Now change the status date to the last date and observer that the CPI
and SPI calculations are correct and both show as 1.

Move the subtask to the top of the list of tasks, so that it is the first
task and not a sub-task any longer. You will notice the CPI and SPI values
are now wrong.

If you have a large project, with many levels and you do this to re-organise
the project, the CPI and SPI values are useless and cannot be used for EVM.
This problem also corrupts the EVM Olap cube extension data.

The obvious solution to re-baseline the project does !NOT! work becuase the
re-baseline algorithm takes the actuals that have been entered and uses them
for the new baseline. If you do this, you loose your baseline infomation
completely, for tasks that contain actual values.

Please help - I think MS might need to produce a bug fix, but I am not sure...

This problem is preventing my company (a very large IT consulting company)
from using MS Project and Project server for Earned Value Management.

Re: Baseline Corruption by JackD

JackD
Fri Feb 17 01:43:00 CST 2006

Stuart,
The simple work around is don't go dragging and dropping tasks.
If you want to rearrange tasks I usually use the grouping feature.
You might also test to see if sorting produces the same result - I don't
expect it would, but if it did, that would be bad.

--
-Jack ... For Microsoft Project information and macro examples visit
http://masamiki.com/project
or http://zo-d.com/blog/index.html
.
"Stuart Penning" <Stuart Penning@discussions.microsoft.com> wrote in message
news:1DFBA26F-BEE1-4292-97C9-49BDBB8BC4DE@microsoft.com...
> I have found that baseline data gets corrupted when the tasks are moved in
> schedules that have been baselined.
>
> To reproduce the problem, do this in a new project:
>
> 1. Create 1 summary task, having one sub task (1 day will do).
> 2. Create a resource with a cost per hour ($1 will do) and assign that
> resource to the sub-task.
> 3. Baseline the whole project and save.
> 4. Set the project status date to the day before the summary task start
date
> (EVM calcs are done based on this date).
>
> Now, switch to the EVM table in the Gantt chart view and add the CPI and
SPI
> columns. The CPI and SPI will be 0 on summary and sub-task - this is
correct.
>
> Open the bottom pane so that you have a combination view with the top
being
> the Gantt chart with the EVM table and the bottom being the standard task
> usage view. Add the actual work row to the task usage pane. Now you should
be
> able to see 8 hours work and a blank for actual work, on the day for the
> subtask.
>
> Select the sub-task. Enter an amount of hours equal to the work value,
into
> the actual work row, on the expected date for the subtask (in the task
usage
> pane). Now change the status date to the last date and observer that the
CPI
> and SPI calculations are correct and both show as 1.
>
> Move the subtask to the top of the list of tasks, so that it is the first
> task and not a sub-task any longer. You will notice the CPI and SPI values
> are now wrong.
>
> If you have a large project, with many levels and you do this to
re-organise
> the project, the CPI and SPI values are useless and cannot be used for
EVM.
> This problem also corrupts the EVM Olap cube extension data.
>
> The obvious solution to re-baseline the project does !NOT! work becuase
the
> re-baseline algorithm takes the actuals that have been entered and uses
them
> for the new baseline. If you do this, you loose your baseline infomation
> completely, for tasks that contain actual values.
>
> Please help - I think MS might need to produce a bug fix, but I am not
sure...
>
> This problem is preventing my company (a very large IT consulting company)
> from using MS Project and Project server for Earned Value Management.



Re: Baseline Corruption by StuartPenning

StuartPenning
Fri Feb 17 16:02:28 CST 2006

Jack,

Thank you for your reply but this is not a solution and it is not an
acceptable work-around either.

I have 10 project managers using the product, all connecting to project
server managing 200 people and this number is growing.

We cannot limit the project manager by restricting the use of this feature
(drag and drop of tasks) because it would make the management of the schedule
an impossible task.

The fact is that people have to use summary and sub-tasks. Another fact is
that the structure of the schedule changes as time goes by.

In the mean time, I have found a manual way to fix the problem, but using
the manual apprach is not feasible for projects that have more than a few
summary tasks - so, not I have to write a macro that does it for me.

The point is that this is a BUG and I can prove it...

"JackD" wrote:

> Stuart,
> The simple work around is don't go dragging and dropping tasks.
> If you want to rearrange tasks I usually use the grouping feature.
> You might also test to see if sorting produces the same result - I don't
> expect it would, but if it did, that would be bad.
>
> --
> -Jack ... For Microsoft Project information and macro examples visit
> http://masamiki.com/project
> or http://zo-d.com/blog/index.html
> ..
> "Stuart Penning" <Stuart Penning@discussions.microsoft.com> wrote in message
> news:1DFBA26F-BEE1-4292-97C9-49BDBB8BC4DE@microsoft.com...
> > I have found that baseline data gets corrupted when the tasks are moved in
> > schedules that have been baselined.
> >
> > To reproduce the problem, do this in a new project:
> >
> > 1. Create 1 summary task, having one sub task (1 day will do).
> > 2. Create a resource with a cost per hour ($1 will do) and assign that
> > resource to the sub-task.
> > 3. Baseline the whole project and save.
> > 4. Set the project status date to the day before the summary task start
> date
> > (EVM calcs are done based on this date).
> >
> > Now, switch to the EVM table in the Gantt chart view and add the CPI and
> SPI
> > columns. The CPI and SPI will be 0 on summary and sub-task - this is
> correct.
> >
> > Open the bottom pane so that you have a combination view with the top
> being
> > the Gantt chart with the EVM table and the bottom being the standard task
> > usage view. Add the actual work row to the task usage pane. Now you should
> be
> > able to see 8 hours work and a blank for actual work, on the day for the
> > subtask.
> >
> > Select the sub-task. Enter an amount of hours equal to the work value,
> into
> > the actual work row, on the expected date for the subtask (in the task
> usage
> > pane). Now change the status date to the last date and observer that the
> CPI
> > and SPI calculations are correct and both show as 1.
> >
> > Move the subtask to the top of the list of tasks, so that it is the first
> > task and not a sub-task any longer. You will notice the CPI and SPI values
> > are now wrong.
> >
> > If you have a large project, with many levels and you do this to
> re-organise
> > the project, the CPI and SPI values are useless and cannot be used for
> EVM.
> > This problem also corrupts the EVM Olap cube extension data.
> >
> > The obvious solution to re-baseline the project does !NOT! work becuase
> the
> > re-baseline algorithm takes the actuals that have been entered and uses
> them
> > for the new baseline. If you do this, you loose your baseline infomation
> > completely, for tasks that contain actual values.
> >
> > Please help - I think MS might need to produce a bug fix, but I am not
> sure...
> >
> > This problem is preventing my company (a very large IT consulting company)
> > from using MS Project and Project server for Earned Value Management.
>
>
>

Re: Baseline Corruption by JackD

JackD
Sat Feb 18 00:26:05 CST 2006

Stuart,

It does not matter to me if it is a bug or not. It is something that must be
lived with or worked around. If writing a macro is the best solution for you
then go for it. You don't need to prove to me that it is a bug. I was just
suggesting a couple of alternatives for you, not arguing that project is bug
free. Only a fool would take on that task.

--
-Jack ... For Microsoft Project information and macro examples visit
http://masamiki.com/project
or http://zo-d.com/blog/index.html
.
"Stuart Penning" <StuartPenning@discussions.microsoft.com> wrote in message
news:F2ED3A69-22D4-4B0D-B0B9-2BA32373911C@microsoft.com...
> Jack,
>
> Thank you for your reply but this is not a solution and it is not an
> acceptable work-around either.
>
> I have 10 project managers using the product, all connecting to project
> server managing 200 people and this number is growing.
>
> We cannot limit the project manager by restricting the use of this feature
> (drag and drop of tasks) because it would make the management of the
schedule
> an impossible task.
>
> The fact is that people have to use summary and sub-tasks. Another fact is
> that the structure of the schedule changes as time goes by.
>
> In the mean time, I have found a manual way to fix the problem, but using
> the manual apprach is not feasible for projects that have more than a few
> summary tasks - so, not I have to write a macro that does it for me.
>
> The point is that this is a BUG and I can prove it...
>
> "JackD" wrote:
>
> > Stuart,
> > The simple work around is don't go dragging and dropping tasks.
> > If you want to rearrange tasks I usually use the grouping feature.
> > You might also test to see if sorting produces the same result - I don't
> > expect it would, but if it did, that would be bad.
> >
> > --
> > -Jack ... For Microsoft Project information and macro examples visit
> > http://masamiki.com/project
> > or http://zo-d.com/blog/index.html
> > ..
> > "Stuart Penning" <Stuart Penning@discussions.microsoft.com> wrote in
message
> > news:1DFBA26F-BEE1-4292-97C9-49BDBB8BC4DE@microsoft.com...
> > > I have found that baseline data gets corrupted when the tasks are
moved in
> > > schedules that have been baselined.
> > >
> > > To reproduce the problem, do this in a new project:
> > >
> > > 1. Create 1 summary task, having one sub task (1 day will do).
> > > 2. Create a resource with a cost per hour ($1 will do) and assign that
> > > resource to the sub-task.
> > > 3. Baseline the whole project and save.
> > > 4. Set the project status date to the day before the summary task
start
> > date
> > > (EVM calcs are done based on this date).
> > >
> > > Now, switch to the EVM table in the Gantt chart view and add the CPI
and
> > SPI
> > > columns. The CPI and SPI will be 0 on summary and sub-task - this is
> > correct.
> > >
> > > Open the bottom pane so that you have a combination view with the top
> > being
> > > the Gantt chart with the EVM table and the bottom being the standard
task
> > > usage view. Add the actual work row to the task usage pane. Now you
should
> > be
> > > able to see 8 hours work and a blank for actual work, on the day for
the
> > > subtask.
> > >
> > > Select the sub-task. Enter an amount of hours equal to the work value,
> > into
> > > the actual work row, on the expected date for the subtask (in the task
> > usage
> > > pane). Now change the status date to the last date and observer that
the
> > CPI
> > > and SPI calculations are correct and both show as 1.
> > >
> > > Move the subtask to the top of the list of tasks, so that it is the
first
> > > task and not a sub-task any longer. You will notice the CPI and SPI
values
> > > are now wrong.
> > >
> > > If you have a large project, with many levels and you do this to
> > re-organise
> > > the project, the CPI and SPI values are useless and cannot be used for
> > EVM.
> > > This problem also corrupts the EVM Olap cube extension data.
> > >
> > > The obvious solution to re-baseline the project does !NOT! work
becuase
> > the
> > > re-baseline algorithm takes the actuals that have been entered and
uses
> > them
> > > for the new baseline. If you do this, you loose your baseline
infomation
> > > completely, for tasks that contain actual values.
> > >
> > > Please help - I think MS might need to produce a bug fix, but I am not
> > sure...
> > >
> > > This problem is preventing my company (a very large IT consulting
company)
> > > from using MS Project and Project server for Earned Value Management.
> >
> >
> >



Re: Baseline Corruption by Mike

Mike
Sat Feb 18 16:02:24 CST 2006

Hi Stuart,

Whether this is a bug or not, the fact that you move a subtask in that
manner turns the original summary into a real task with 0 hours work but
with 8 hours of baseline work that Project has to account for. I suggest
this sudden extra task will have an effect on the original data as it wasn't
there before! What effect that will have I'm not prepared to spend my time
investigating! It seems a most irrational thing to do! Have you tried
deleting the original summary task it now being redundant?

Incidentally, as you're using Project Server, next time please post on the
server newsgroup. Please see FAQ Item: 24. Project Newsgroups. FAQs,
companion products and other useful Project information can be seen at this
web address: http://project.mvps.org/faqs.htm

Mike Glen
Project MVP








Stuart Penning wrote:
> I have found that baseline data gets corrupted when the tasks are
> moved in schedules that have been baselined.
>
> To reproduce the problem, do this in a new project:
>
> 1. Create 1 summary task, having one sub task (1 day will do).
> 2. Create a resource with a cost per hour ($1 will do) and assign that
> resource to the sub-task.
> 3. Baseline the whole project and save.
> 4. Set the project status date to the day before the summary task
> start date (EVM calcs are done based on this date).
>
> Now, switch to the EVM table in the Gantt chart view and add the CPI
> and SPI columns. The CPI and SPI will be 0 on summary and sub-task -
> this is correct.
>
> Open the bottom pane so that you have a combination view with the top
> being the Gantt chart with the EVM table and the bottom being the
> standard task usage view. Add the actual work row to the task usage
> pane. Now you should be able to see 8 hours work and a blank for
> actual work, on the day for the subtask.
>
> Select the sub-task. Enter an amount of hours equal to the work
> value, into the actual work row, on the expected date for the
> subtask (in the task usage pane). Now change the status date to the
> last date and observer that the CPI and SPI calculations are correct
> and both show as 1.
>
> Move the subtask to the top of the list of tasks, so that it is the
> first task and not a sub-task any longer. You will notice the CPI and
> SPI values are now wrong.
>
> If you have a large project, with many levels and you do this to
> re-organise the project, the CPI and SPI values are useless and
> cannot be used for EVM. This problem also corrupts the EVM Olap cube
> extension data.
>
> The obvious solution to re-baseline the project does !NOT! work
> becuase the re-baseline algorithm takes the actuals that have been
> entered and uses them for the new baseline. If you do this, you loose
> your baseline infomation completely, for tasks that contain actual
> values.
>
> Please help - I think MS might need to produce a bug fix, but I am
> not sure...
>
> This problem is preventing my company (a very large IT consulting
> company) from using MS Project and Project server for Earned Value
> Management.