WHEN:Nov 24-25

WHERE: Belgium

MORE INFO:

http://www.eva-europe.eu/2010/ 

 

NASA PM Challenge
WHEN: Feb 9-10

WHERE: Long Beach CA

MORE INFO: http://pmchallenge.gsfc.nasa.gov/index.htm

 

Book Review. "Measuring Time, Improving Project Performance Using Earned Value Management", Mario Vanhoucke. Volume 136 in the International Series in Operations Research and Management Science, Springer Press, ISBN 978-1-4419-1013-4, 155 pages, hardcover.
 
Within the last ten years three emerging concepts for the use of EVM data in estimating project duration (and thus project's estimated completion date) were proposed: Planned Value method, Earned Duration, and Earned Schedule. Mario Vanhoucke, in concert with Stephan Vandevoorde and others, performed academic research on these concepts using complex project modeling and Monte Carlo methods. This book is a summary of the research effort.
 
The book is not a text book on EVM. It is a thorough discussion about their research methods and analysis of their results. This book provides the reader insight into how the research was conducted, but also when each of the three competing project duration algorithms appears to work best. It also provides a useful concepts for others considering research in the analysis of project performance using numerous and complex models of project activity networks.
 
We know from experience that while BCWP helps shows progress, it's not sensitive to the critical path and which work has been completed. One can have lots of BCWP from doing the work and the wrong time. The book also researches the "P-factor" index. This index (0-1.0), also called "Schedule Adherence", illustrates how much of the cumulative BCWP (EV) earned to date is per the project network compared to the total BCWP.
 
Most EVM research has used limited sets of real world project data or project models. The outcomes of such research was always influenced by the projects' characteristics such a size, type, industry, complexity, etc. One might also suspect that the project chosen for the research were subconsciously selected to favor the research. This often made the results of such research "not applicable on my projects."
 
The author also begins with the application of the three methods to three real projects. But what makes Vanhoucke's research unique is his use of computer generated networks. Using four parameters he is able to model the topography of various activity networks. He then assigns planned durations and a probability distribution to each of the activities. Using Monte Carlo methods each activity and project network is executed and EVM data collected after each project reporting interval. These data form the basis of his research into each of the methods.
 
His conclusion is that Earned Schedule outperforms the other methods for most projects when schedule performance is accurately reported throughout the project duration. The use of P-factor to create a more accurate "effective" BCWP does not appear to improve duration prediction over use of traditional BCWP.
 
This book also discusses research on how one individual activity might influence the project's outcome and the value of top down versus bottom up project tracking.
 
This book is not required for an understanding of Planned Value, Earned Duration, and Earned Schedule estimating methods. But if the goal is understand the merits of each method, P-factor, and "effective" earned value, then this book is worth reading.