OTB reprogram S=P or S=P=A

Hi Andrew

Welcome to Planning Planet

I would really like to help but I have no idea what you are talking about.

Best regards

Mike Testro

Whew! Thanks, Mike! Thought it was just me...

Fraternally in project management,

Steve the Bajan

Ah sorry everyone. I've been speaking internal Earned Value Management colloquiallism for a few days, guess my brain is wired. Let me try again...

A project I'm looking at is running way over budget and schedule, ie Over Target Baseline (OTB), so they have decided to replan the rest of the project, and acquire further budget, in order to set a new more useful performance baseline. As normal for Earned Value Management rebaselining, they they need to reset the Schedule Performance Index to zero by setting BCWS equal to BCWP (S=P), for the current work. We are trying to convice them to also reset the Cost Performance Index by setting BCWS=BCWP=ACWP (S=P=A), but they are reluctant.

Pretty spefic EVM stuff, does that make it any more clearer?

Hi, Andrew. Yes, that makes it much clearer – I now know exactly what I don’t understand! It’s the phrase “for the current work”. Does that mean the work scheduled to this point (BCWS), the work completed so far (BCWP), the ongoing work (i.e., work that is started but not finished), or the future work (not yet started)? Or all of them?

In general, I am opposed to re-baselining except for product scope changes. If more money/time is needed, we should just accept that and make it available – there really is no need to change the earned value baseline. However, team members often feel pressure when running late or going over budget, and this can lead to bad decisions – out-of-sequence work, scope and quality pruning, risky shortcuts, cheaper and less skilled resources, etc. To prevent that feeling of pressure, organizations will sometimes change the earned value baseline to reflect additional funding when it becomes clear from the CPI and SPI that costs and durations have been underestimated. However:

• The original baseline plan should always be maintained as an important artifact for the project postmortem and ABCP analysis. That is the only way we are ever likely to improve our estimating for future projects.
• Baseline modification (again with the exception of scope change) should not, in my humble opinion, ever be done for “sunk work”– if we performed the first half of the project at an SPI of .90 and a CPI of .80, so be it: “Lots of water under the bridge, Miss Ilsa.”
• If we are going to provide additional resources for future work AND we want a new baseline plan (two different things!), we should take a snapshot of where we are, with the new CPI to be based on the new BCWS.
• But note that there may be occasions when the additional resources may make our original duration estimates now achievable. If this is so, the activity schedule can just be “lifted” and place back down after the current data date, with just the earned value weights of each activity increased by the additional funding.
• Under any circumstances, we now essentially have a new project with a new BCWS against which we should now start tracking afresh. Sunk work and sunk costs are sunk unless your project is to develop a time machine!

Andrew, I recently completed my new book Managing Projects as Investments: Earned Value to Business Value, due out from CRC Press in September. It has two chapters on earned value of which I’m quite proud. I think they explain the concepts in a simple and comprehendible manner, identify the shortcomings and easy ways to improve them, and illustrate some of the more sophisticated techniques. So I am going to kinda plagiarize from my own book and share a couple of concepts that might help you:

For your new baseline of the future work, I urge saving a separate BCWS for schedule (SV, SPI), distinct from the one for cost (CV, CPI). Even though traditional EVM pretends they aren’t, schedule and cost are different! Schedule is driven by the critical path, but the cost is critical path blind: a dollar is a dollar is a dollar, whether spent on an activity with 30 days of drag or 300 days of float. The critical path is THE most important aspect of schedule, so to ignore it in tracking schedule, as traditional EVM does, is hugely distorting: an activity with 300 days of float but a budget of $100,000 is worth 10 times as much to the SPI as one with 30 days of drag but a budget of only $10,000. This not only distorts the SPI, but leads to “gaming” the SPI by grabbing noncritical but big budget activities and even doing them out-of-sequence to the detriment of the critical path activities and ultimately to the project duration. This can be fixed by:

1. Saving a separate BCWS for schedule where all the dates are ALAP (i.e., the backward pass) where there is no float.
2. Only giving BCWP credit to complete it activities in the reporting period when they are scheduled to occur WITH THE EXCEPTION OF activities on the critical path of the current working schedule i.e., the only activities where finishing early can actually compress the schedule).

I hope this all helps, Andrew. I will now reference my book (from which I stole this last part!). If you think this info has been helpful, I’d certainly appreciate your consideration of a pre-order.

http://www.crcpress.com/product/isbn/9781482212709

Fraternally in project management,

Steve the Bajan