none
How to build an enterprise resource pool for resource constraint analysis. RRS feed

  • Question

  • Hello everyone,

    I am using Project Server 2010. I'm attempting to build an enterprise resource pool so that I can compare resource capacity to resource demand. I have read Implementing and Administering Microsoft Project Server 2010 For Project Managers by Gary L. Chefetz, Dale A. Howard and TonyZink (an excellent resource that I highly recommend) and a series of Microsoft articles beginning with At a Glance: Roles vs. Generic Resources (http://blogs.msdn.com/b/project/archive/2010/10/27/at-a-glance-roles-vs-generic-resources.aspx). I think I understand the concepts, but I have a few questions.

    So far, I have:

    1.      Defined and configured primary resource roles.

    2.      Created resources to represent capacity:

    a.       I created named resources for the ‘real people’ in the company, assigned each resource a max unit of 100% and assigned each resource a primary role.

    b.      I created a generic resource for each role and assigned each resource a max unit based on the number of named resources for its role and used a custom flag to identify that the generic resource is used as a placeholder per Microsoft article http://office.microsoft.com/en-us/project-server-help/create-resources-to-represent-capacity-HA101865477.aspx?CTT=5&origin=HA101865473

    3.      Created resources to represent demand:

    a.       I created a generic resource for each role and assigned each resource a max unit of 100%.

    Resource name

    Type

    Role

    Max Units

    Placeholder

    Paul

    Named

    Developer

    100%

    No

    Ryan

    Named

    Developer

    100%

    No

    Developer-Capacity

    Generic

    Developer

    200%

    Yes

    Developer-Demand

    Generic

    Developer

    100%

    No

    Project Manager-Capacity

    Generic

    Project Manager

    0%

    Yes

    Project Manager-Demand

    Generic

    Project Manager

    100%

    No

     

     

     

     

     

     

    Now finally the questions.

    1.       Is the table above the correct way to build a resource pool that includes 2 developers and a future Project Manager?

    2.       My belief is that Project Server will report capacity for the Developer role as 200%, not 400%. Is that correct?

    3.       Are the Max Units set properly to take advantage of resource constraint analysis?

    4.       Assuming a 40 hour work week and a resource plan that called for 40 hours of work for the Project Manager role, would Project Server show an over allocation for the Project Manager role because there is no named resource with the Project Manager role assigned? The goal is to show that I need to hire a Project Manager to complete the project.

    Thanks

    Saturday, December 11, 2010 6:17 AM

Answers

  • I looked at that second article, which seems to detail your approach pretty well.  Personally, I go for a slightly simpler method.  This approach works for me, although it has it's drawbacks on occasion. Here's the way I typically configure this sort of thing:

    1) Set up all of my named resources and map to a specific role.  Typically, in an IT shop, I use a MaxUnits of 80% or so.  Depending on the organizational process, I'll set the named resources to have a default booking type of Committed.

    2) Set up one generic resource for each role.  I set the Max Units for the generic role to 80% as well.  Optionally set the generic resources to a default booking type of Proposed.  I just use the role name as the name of the generic resource.

    That's pretty much it from a setup/configuration standpoint.  From my perspective:

    1) Setting a generic resource Max Units to something like 400% makes it hard to develop good templates, as those templates won't display whether or not the generic resource is overallocated - until you've allocated more than 400% at any one time.

    2) Demand includes both those projects that have been proposed and those projects that have been committed.  The important thing then is to subdivide the demand out into those two categories.

    I'm curious what other folks have to say...


    Andrew Lavinsky [MVP] Twitter: @alavinsky Blog: http://blogs.catapultsystems.com/epm
    Saturday, December 11, 2010 1:34 PM
    Moderator

All replies

  • I looked at that second article, which seems to detail your approach pretty well.  Personally, I go for a slightly simpler method.  This approach works for me, although it has it's drawbacks on occasion. Here's the way I typically configure this sort of thing:

    1) Set up all of my named resources and map to a specific role.  Typically, in an IT shop, I use a MaxUnits of 80% or so.  Depending on the organizational process, I'll set the named resources to have a default booking type of Committed.

    2) Set up one generic resource for each role.  I set the Max Units for the generic role to 80% as well.  Optionally set the generic resources to a default booking type of Proposed.  I just use the role name as the name of the generic resource.

    That's pretty much it from a setup/configuration standpoint.  From my perspective:

    1) Setting a generic resource Max Units to something like 400% makes it hard to develop good templates, as those templates won't display whether or not the generic resource is overallocated - until you've allocated more than 400% at any one time.

    2) Demand includes both those projects that have been proposed and those projects that have been committed.  The important thing then is to subdivide the demand out into those two categories.

    I'm curious what other folks have to say...


    Andrew Lavinsky [MVP] Twitter: @alavinsky Blog: http://blogs.catapultsystems.com/epm
    Saturday, December 11, 2010 1:34 PM
    Moderator
  • Andrew,

    Thanks for the post. I was hoping to hear more from the community on this one. Your approach makes sense and is easier to follow. What drawbacks have you experienced using it?

    I think I’ve got the demand side covered, but I’m still struggling to completely understand just how Project Server determines capacity? Can you point me to a resource that discusses it in detail? I’m Google-ing like a madman, but can’t find anything.

    In my original post,  I have 4 resources with the Developer role assigned. What does that mean to Project Server? Does it add the max units of all the resources to determine capacity, or does it exclude generic resources in capacity computations? How does Project Server compute resource capacity?

    Thanks

    Monday, December 13, 2010 1:37 AM
  • As I recall, it sums up both generic and named resources to identify maximum capacity - although I could be off on that.  That's where my technique would have a drawback, as the generic resource could skew your capacity slightly.  Still - it's relatively easy to filter out the generic from named resources - and the Proposed vs Committed gives you one more layer of control.  That should give you a pretty good idea of what's going on.

    The other issue would be if you had two developers in your template for different tasks.  Then you would have to have generic Developer 1 and generic Developer 2 as resources - unless you wanted to just assign the task to a generic Developer at 200%...


    Andrew Lavinsky [MVP] Twitter: @alavinsky Blog: http://blogs.catapultsystems.com/epm
    Monday, December 13, 2010 1:59 AM
    Moderator