Project Management Institute, ISBN: 979-1-935589-35-8,151 pages, soft cover or pdf download, $55.95 (less for PMI members)

PMI® released the 2nd Edition of the Practice Standard for Earned Value Management at the October North American Congress. This replaces the first edition which I had the opportunity to work on, but was sadly disappointed in the final product. (Not my fault.) This edition is a complete and welcomed rewrite and is truly a complete treatment of EVM. While the first edition was focused on "what is EVM" this edition is focused on "how to do EVM". It is process focused and ties to processes in the PMI Guide to Project Management Body of Knowledge (PMI PMBOK). Its chapters address "EVM Overview", "Project Organization", "Assigning Responsibility", "Develop Schedule", "Establishing Budget", "Determine Measurement Methods", "Establish Performance Measurement Baseline", "Analyze Project Performance", and "Maintain Project Performance Baseline".

Appendices address schedule analysis (earned schedule), integration of EVM with risk management, deployment of EVM systems, and EVM pitfalls and recommendations.

It's CPI is greater than the first considering you get 151 pages of good stuff for $56(US) versus the first edition of about $1 per page of useful content.

A well qualified international volunteer team, lead by J. Greg Smith, wrote the Practice Standard. It was reviewed by an equally qualified international volunteer team. In addition public comments were received and adjudicated. The resulting Standard is international and industry neutral. (Differences between US ANSI 748 EVM and the PS are noted throughout.)

The goal of the PS team was "to provide the global community with a readable document that (1) explains the basics of EVM clearly and concisely for new practitioners and (2) provides a reasonable extension ...to those who want more ....than given in the PMBOK Guide." The team achieved their goal.

The previous version was PMI's most popular "Standard" behind the PMI PMBOK. An academic study found this first version to be the mostly widely accepted standard for EVM worldwide. I expect sales and use of this version to exceed that of the first edition and help bring the power of EVM to projects, industries, companies, and governments worldwide.

Back to top

Call them hurricanes, typhoons, or cyclones. Whatever you call them, when they are on TV they get the attention of millions of people. The challenge for the national weather agencies is to predict where, when, and with how much force the storm will make landfall. "Spaghetti lines" are what the weather services call the various wiggly lines that attempt to predict the storm's path from its present position to likely future positions. Same data, different models, different prognostication. Sometimes the lines are nearly the same, sometimes they diverge significantly. What that shows is the range of possible paths.

In EVM we start with the same four data: PV, EV, AC, BAC. We should use these data as input to several models so we can also forecast a range of outcomes. No estimate will be right, but if we have a several estimates we can see how closely they agree or don't agree. This can show us a range of possible outcomes. The effort to put EVM in place and collect performance data has been done by the time we have our EVM data. Why not run the data through all the "models". It's just a formula and computers do math really well. Here are a few "models" that can be used with our basic EVM data.

• EAC=BAC/CPI
• EAC = AC_cum + (BAC-EV)/(CPI*SPI)
• EAC = AC_cum + (BAC-EV)/(CPI_n {for the last n months})
• EAC = AC_cum + (BAC-EV)
• ECD (Estimated Completion Date) = Planned Duration/(SPI{t})

• Statistical forecasts of final CPI and SPI based upon values to date, periods remaining, and desired confidence level (percent)

Other non-EVM models:

• EAC = AC_cum + Bottom-up estimate
• ECD from Critical Path Analysis

Only by applying a number of models and methods can we know the range of possible outcomes. Sharing this with our stakeholders helps them realize the future is uncertain and no single point estimate will be correct.

As the hurricane turns and heads in a new direction the weather experts will state, "Yeah, one of our models said this might happen." You should strive to be able to make the same statement about your project.

Back to top