Baselining vs realistic duration vs actual work vs resource overal

B

Bert

How would you properly baseline a project when the overall task duration will
be larger than the actual work any task requires. (Note: Many resource's
other job demands will seldom allow tasks to be started when scheduled. 1 to
5+ day delay)

To avoid overallocation of resources will require project managers to set
unrealistic baslines for their respective projects. (Note: Senior Management
wants all projects to be baselined.)

Can anyone provide some sage advice or best practices used to overcome this
problem?
Thanks
Bert
 
D

Dale Howard [MVP]

Bert --

I don't think the problem isn't how to baseline your projects. I believe
the problem is how you PLAN your projects initially. From your description,
it sounds like your projects typically slip because resources start late on
tasks due to other demands. If this is the primary reason for your project
slipping, then I believe you need to plan your projects with this kind if
slippage in mind. One way to handle this would be to make sure you include
a Milestone task at the end of every Phase in your project, and probably at
the end of every deliverable section as well.

For example, suppose that you add a Milestone at the end of every
Deliverable section. In the first Deliverable section, your past historical
record shows that the tasks in this section typically finish 5 days later
than originally scheduled. How can you "build in" this potential slippage
initially, so that your problem has "wiggle room" for things to go wrong?
Here's how: add a Start No Earlier Than constraint on the Milestone task
and set the Constraint date five days later than the current Finish date for
the Milestone. This creates a gap between the last task in the Deliverable
section, and does impact the Critical Path, but it allows for time for
things to go wrong, like resources not starting tasks on time. Using this
technique, you would set a SNET constraint on each Deliverable milestone
task. Another technique would be add SNET constraints to the Milestones for
each Phase section instead. This is a technique that I personally recommend
to my students because it builds slack time into your project, which allows
for things to go wrong, and when things do go wrong (which they will), then
the slack time "absorbs" the slippage. After planning your project using
either of these two techniques, baseline your project.

Please note that the above is only my opinion, and is only one way to
resolve this problem. I hope that the others in this group will share some
ideas as well. Hope this helps.
 
B

Bert

Your analysis of our current project planning issues are right on.
Excellent solution, very effective for my situation too.

If anyone has other options to handle this problem, I would like to hear
them as well. (I have a core project team who will want to evaluate all
options available)
Thanks for your time
Bert
 
B

Ben Howard

Hi Bert, my 2p worth...

Accurate planning in your case is always going to be difficult given that it
appears that the resources have many masters, and projects obviously have a
low priorty until things become "critical". The pragmatic method that Dale
recommends deals with your current situation, but setting constraint dates
limits the flexibility of the scheduling engine within MS project. Given
that management are interested in measuring performance then you need, as an
organisation, to give some thought on how you solve this business issue.
Also, from a tool perspective, you could set your resources at < 100%, (say
50% of their time is allocated to project work), this will retain the
automatic scheduling whilst allowing resources to work on other things.
Obviously, the % of project work to other work needs to be agreed with
management etc.

HTH
--
Thanks, Ben.
http://appleparkltd.spaces.live.com/
 
R

rt

Hi Bert, my 2p worth...

Accurate planning in your case is always going to be difficult given that it
appears that the resources have many masters, and projects obviously have a
low priorty until things become "critical". The pragmatic method that Dale
recommends deals with your current situation, but setting constraint dates
limits the flexibility of the scheduling engine within MS project. Given
that management are interested in measuring performance then you need, as an
organisation, to give some thought on how you solve this business issue.
Also, from a tool perspective, you could set your resources at < 100%, (say
50% of their time is allocated to project work), this will retain the
automatic scheduling whilst allowing resources to work on other things.
Obviously, the % of project work to other work needs to be agreed with
management etc.

HTH
--
Thanks, Ben.http://appleparkltd.spaces.live.com/








- Tekst uit oorspronkelijk bericht weergeven -
Click to expand...

I agree with Ben. Adding constraints will make managing the schedule
all the more time consuming. And for what, I wonder. If all project
plans slip all of the time, don't cover that up with clever tricks -
solve the root problem!
 
G

Gary L. Chefetz [MVP]

True when the constraints are inflexible. SNET constraints are benign unless
you are trying to determine an early finish for the project. In Bert's
situation, if he uses Project Tracking, his schedule will become loaded with
them anyway!

--

Gary L. Chefetz, MVP
MSProjectExperts
For Project Server Consulting: http://www.msprojectexperts.com
For Project Server FAQS: http://www.projectserverexperts.com


Ben Howard said:
Hi Bert, my 2p worth...
Click to expand...

but setting constraint dates
limits the flexibility of the scheduling engine within MS project
 
B

Bert

Thanks for everyone's input: One more question please, what would the
effects or impact be from simply deselecting the "Effort Driven" button have
on the scheduling output as another option for managing this?
 
D

Dale Howard [MVP]

Bert --

Deselecting that option would have absolutely no impact on your problem.
The "Effort Driven" option only comes into play when you add "helpers" to a
task after initially assigning at least one resource. On an Effort Driven
task, when you add one or more "helpers" to the task, the system distributes
the Remaining Work proportionately to each resource and recalculates the
Duration accordingly. So, if that is the case, how could that possibly
impact your situation? Hope this helps.
 
Top