Predicting the Future

From the Voices on Project Management Blog
by , , , , , , , , , , , , , , ,
Voices on Project Management offers insights, tips, advice and personal stories from project managers in different regions and industries. The goal is to get you thinking, and spark a discussion. So, if you read something that you agree with - or even disagree with - leave a comment.

About this Blog

RSS

View Posts By:

Cameron McGaughy
Marian Haus
Lynda Bourne
Lung-Hung Chou
Bernadine Douglas
Kevin Korterud
Conrado Morlan
Peter Tarhanidis
Mario Trentim
Jen Skrabak
David Wakeman
Roberto Toledo
Cecilia Wong
Vivek Prakash
Cyndee Miller
Shobhna Raghupathy

Recent Posts

Level 5 Leadership: Taking Your Project from Good to Great

Sprinting a Marathon

How Managers Can Grow Into Leaders

Managing The Last 100 Feet

The Network Diagram Mentality

Email Notifications off: Turn on


My earlier post, Our Biggest Unused Weapon, posited that the ability of even a basic earned value management system (EVMS) to predict when a project will be complete and how much it will cost at completion is unparalleled in the management information system world. But I also argued it was being under-used by most practitioners.

William Goelkel, PMP, responded with:

"EVM can lead us to make bad decisions, or forecasts, when the standard against which we measure everything is the plan we baselined at the point in the project's life cycle when we were most ignorant of its requirements."

And

"I'm not convinced a calculated EAC [estimate at completion] as you described is always useful."

I'd like to further my arguments to the contrary while responding to Mr. Goelkel's thoughtful comment.

Mr. Goelkel's main example concerns software projects, where "goals change and our understanding of the users' needs evolves." Either this understanding evolution is captured formally and introduced into the baseline, or it is not. If it is, then there's no problem. If it is not, then we have scope creep, the most pernicious attribute of managing projects.

Even so, EVMS have a remarkable ability to predict the future, even when the baseline has been turned to rubber.

Say you had a US$100,000 project, but "evolving" customer expectations will end up costing the contractor double that, or US$200,000. At the point that the project should be half-done (cumulative budget = US$50,000, and actual costs are at the same level) the task leader assesses that she is only 25 percent done. The calculated EAC will instantly indicate the real EAC of US$200,000, allowing for either elimination of the scope creep or a request for more money.

For a more real-world example, try to find a project that (impishly) has a "Project Managers' EAC" column right next to the calculated EAC in their cost performance reports.

Ninety-nine times out of 100, the project manager's version is lower than the calculated one and less accurate. It's like clockwork.

In my next post, I'd like to tackle how project management trends in the software world--like scrum or Agile--are actually attempts to accommodate the rampant scope creep that so often afflicts those projects.

For now, I'd like to hear what other EV practitioners have to say. I'd also like to thank William Goelkel for this discussion.

Posted by MICHAEL HATFIELD on: April 06, 2009 12:03 PM | Permalink

Comments

Glen B. Alleman
Michael, If you're rubberbanding the baseline, the EV metrics are worthless. The fundamental tennet of EV - see the DoD guidelines - is a stable baseline. Rebaselining requires a re-evaluation of all future work activities, and thier impact on ETC. The "remarkable ability" of EV must be tempered with caution in the presence of changing requirements, less than firm measures of "done," and the "capacity for work assessments," that are commonly missing from software development projects. Glen B. Alleman VP, Program Planning and Controls Aerospace and Defense Denver Colorado www.lewisandfowler.com

Dr. Azhar M. Khan
EVM only provides an estimate and a guideline and is not a hard and fast measurement. It does not consider contingencies or cultures. In many developing countries the labor (people) working on projects overstay leaves for weeks on important festivals and project work virtually comes to a stand-still. Real world is non-linear and unpredictable and formulas like EAC=BAC/CPI actually do not work. Yet EVM is a great way of getting a first order estimates of the things.

Ray W. Stratton, PMP, EVP
EVM can be applied in many ways, other the ANSI 748. But even the formal ANSI provides for "Planning Packages" and "Summary Level Planning Packages" as a way to leave some of the detail plans for downstream work to a point when detail is really known about the nature of the work. A stable baseline is needed for the stakeholders, even if EVM is not used. But, re-planning should take place when (1) new information comes to light about the nature of the work, OR (2) when there is no hope of recovery to the current baseline. We should not plan in detail, work that we do not understand in detail. We (should) learn something new about the project every day. To make use of this I recommend periodic (say quarterly) reviews and UPDATE of the project schedule and cost estimates. EVM data can help in this analysis and determine if the new plan is remotely achieveable (TCPI vs CPI, earned schedule forcast completion dates, etc.), and ANSI allows revisions, just as long as its approved by the right people.

Maureen Gan
Dear Michael I am glad that someone touched on this topic, as it was not an easy one here. Thanks for letting us have your thoughts on "predicting the future" through a more quantifiable means on the project schedules and project costs. Although grasping the concepts need some foundations on EVA, the practical applications could be improved. I find that the formulaes make sense to PM professionals and we need to translate this to a common language where business and sponsors/stakeholders could understand. Could you share with us whether a separate project accounting personnel would be more suited for computing ETC, EAC, AC values or would someone from the PMO is able to track these better ? I think that the discussions on having a project baseline for scope,costs and time is important for us to start with. I am supporting these concepts and I think that it really gives visibility of the numbers and variances that tells the story of whether the project is running late, on-time and whether there is any cost overrun issues. Regards Maureen Gan PMP

Rita Coppola
I am a certified Project Manager in the construction field. I am very interested in using EVM for my projects. There are so many formulas and I am wondering if someone can tell me which EVM fomulas I should concentrate on using. I am interetsed in providing this data not only for my use and the use of my project managers, but also to use in high level presentations. Thank you

Mark Lester Dy
I agree with Glen. If there are changes to the plan (and assumed to have gone through integrated change control and approved), metrics shall be updated as well as the EV values and calculations. Mark Dy, PMP Program and Project Manager Globe Telecom Inc Philippines

Rodrigo Buzeta
Hi Michael, I think EVM is very powerful. I always use it but It is not always efective. The application of EVM to projects that have a big effort that may be categorized as “non-schedule-basedâ€쳌 could be considered ineffective. “Non-schedule-basedâ€쳌 efforts include services composed primarily of Level of Effort (LOE) activity, such as support services or continuous services. The other case where EVM is not always applicable is when the project has a big level of risk. When the monetary value of expected risk is significative compared with PV then it is very dificult to take decisions from the EVMS. Rodrigo Buzeta - PMP, PgMP Synapsis

N. Gopi Krishna
EVM is undoubtedly a powerfull concept in that, it allows you to quantify the project status and also express it is monetary terms. But the EAC has a small drawback. Mathematically, it looks at the CV to be linear and extrapolates the EAC on a similar scale. (Irrespective or variances being typical or atypical). This is not such a straight forward scenario in real world projects. If the design phase has a delay because of difficult requirements OR unfamiliar technology etc, implementation can be slowed much more than a simple linear estimation. Also, current EAC formauls (atleast those in PMBOK) do not help in cases which had both typical and atypical variances. A more carefull mathematical simulation/formulas will be needed in this case. That said, since PMBOK does not prohibit you to use a different formula for EAC, if one can substantiate his formula's logic, I think EAC and EVM help tremendously in giving a realistic picture of what the costs will be at the end of the day. This can decide a lot of things, not in the least the project's success and customer satisfaction.

Michael Hatfield
Wow! I'm glad to see this level of discourse! Rather than address the concerns wholesale, I'm going to write another blog this coming week that furthers my points. Thanks for reading. Oh, one more thing: Ray, you're going to have to buy the first round in Naples at CPM's EV World Seminar, just for the time and effort it's gonna take me to get you to see things my way! Regards, Michael

Anonymous
After reading prior comments, I have gathered EVM not to be effective unless there is a strict time line component. I assume that the present value of money and future value of money, which is a terminology from another industry, but relevant in the use of the EVM formulas, is at play.

You need to know the amount of time passing to ascertain a value at a point in time (absorption rate versus units). That being said, could anyone tell me the relationship between the end-product of the EVM process and a pro forma, as produced in real-estate and construction?

Historically, I have seen pro formas used for assessing the feasibility of projects with input of all the known -- at a conceptual phase -- financial factors playing out over the time of the project. It is a technique used to build a business case.

EVM seems to be the same thing, only during the actual implementation stages, with generally known processes and units. Anyone care to comment? What are the likenesses, what are the differences? Thanks.

Editors’ note: on June 8, a few commenters’ names were lost due to human error. We deeply apologize.

Please Login/Register to leave a comment.

ADVERTISEMENTS

You can say any foolish thing to a dog, and the dog will give you a look that says, "My God, you're right! I never would've thought of that!

- Dave Barry

ADVERTISEMENT

Sponsors

>