How can an EV be calculated when there are different versions of a project plan?

Question

There is a project manager who is trying to come up with a way to calculate EV under a specific situation where there is a project management system which allows different versions of project plans.

The PMs starts with a initial project plan, cost are allocated to different tasks across different period, at the end of each month, PMs review their project plan and revise the plan according, this includes revising the project structure (adding/removing tasks) and project costs. During the running of the project, various cost are also charged to the project.

Here is an example of the project:

Revision 0 (cost/schedule plan): 

month  |  1  |  2 
task-1 | 100 | 200 
task-2 | 200 | 100 

Revision 1 (revised at the end of month 1) 
month  |  1  |  2
task 1 | 200 | 200 
task 2 | 200 | 100 

Revision 2 (revised at end of month 2) 
month  |  1  |  2
task 1 | 200 | 300 
task 2 | 200 | 100 

Actual Project Cost:

month  |  1  |  2 
task-1 | 200 | 300 
task-2 | 200 | 100 

In this case, how can the PM modify the formula (shown below) to calculate the EV in this situation? Is there a better formula to use in this situation?

Here is the formula the PM is working with:

PV= PV of previous plan 
EV= (actual cost Todate)/(total cost of new plan) * (PV according to previous plan) 
AC = actual cost to-date. 

Answer 1

What you are describing here is a complete violation of the procedures to do EV. The baseline of yoru schedule and costs cannot change in this manner. There are specific reasons for and methods to change the baseline but it can never be changed in the manner this PM describes.

You can strike multiple baselines and do EV monitoring against each baseline. But you wouldn't mix and match data across baselines. I have never monitored multiple baselines and am not really sure when one would want to do that. Maybe someone else can think of a scenario where that would be appropriate. So, if this is what the PM is trying to do, then each variable--PV, EV, and AC--would be captured separately for each baseline. Each baseline would have its own PV and EV and AC data. The total AC data might equal summed up; however, it would be applied against the work packages differently depending on how the work was decomposed, the WBS and activities, for each baseline.

Answer 2

I would agree with David. What you're describing isn't EV, but simply monitoring actuals, and doing a cost-to-complete calc each month. Whenever you adjust things in this way you're eliminating the purpose of EV.

In this case, I would suggest discarding the EV processes, and simply use a percent complete calc, and projected cost to complete. Both of these can be updated fairly quickly each month, and adjusted whenever tasks or costs are modified in the manner you're describing. EV is just an unnecessary step in this situation.