Assessing Earned Value Management and Earned Schedule Forecasting


By Walt Lipke

PMI Oklahoma City Chapter

Oklahoma, USA



Recent research indicates cost and schedule forecasting from EVM data is improved when the performance factor, PF = 1, is used. This paper uses a small set of real data to examine the research finding, to either confirm or refute. As well, the application of PF = 1 is employed in statistical forecasting; results are tested and compared to the index method. Observations from the research and this study are made referencing historical studies. Further research is encouraged on these topics, but with some precaution when real data is used.


The 2015 paper, “Empirical Evaluation of Earned Value Management Forecasting Accuracy for Time and Cost” authored by Batselier and Vanhoucke, is the inspiration for this article [Batselier et al, 2015]. Their paper is an impressively comprehensive examination of forecasting from the use of Earned Value Management (EVM) data taken from 51 projects, predominantly construction.

In the history of EVM and Earned Schedule (ES) research, covering 25 years for cost and 15 years for schedule, one type of forecasting formula, incredibly, has been ignored. Included in these past studies are several published by Christensen, Vanhoucke, Crumrine, and Lipke. Uniquely, Batselier and Vanhoucke (B&V) examine several methods of forecasting. B&V demonstrate overwhelmingly in their analysis this ignored formula yields forecasts more often better than the ones most frequently employed by EVM and ES practitioners.

This article, using a smaller set of data than that used by B&V, attempts to corroborate their finding. The primary objective, however, is to implement the improvement shown for deterministic forecasting into statistical forecasting. The focus is to assess whether the improved nominal forecast translates to better statistical forecasts. As well, the investigation may reveal logical reason for the B&V results.

The subsections following, EVM & ES Forecasting, and Statistical Forecasting, provide background for understanding the remainder of the article.

EVM & ES Forecasting

EVM and ES forecasting formulas are very similar. They each have the same basic construct; i.e., the forecast is equal to the current value plus the remainder yet to accomplish divided by a selected performance factor.

Before discussing the formulas, the following EVM and ES terminology is introduced in table 1. It is assumed the reader has a fundamental understanding of EVM and ES. If a more complete description is needed, please reference the following: Practice Standard for Earned Value Management [PMI, 2011], and Earned Schedule [Lipke, 2009-2].

Table 1. EVM and ESM Terminology and Formulas

In the B&V paper several performance factors (PF) are examined for EVM and ES forecasting. For this paper, only four are used. As depicted in table 1, cost applies 1 and CPI, while schedule uses 1 and SPI(t). The reason only these are studied is to corroborate the B&V finding that use of PF = 1 provides in most instances a better forecast of the actual outcome than does the most often used cumulative value for the performance indexes.


To read entire paper, click here


About the Author

Walt Lipke

Oklahoma, USA

Walt Lipke
retired in 2005 as deputy chief of the Software Division at Tinker Air Force Base, where he led the organization to the 1999 SEI/IEEE award for Software Process Achievement. He is the creator of the Earned Schedule technique, which extracts schedule information from earned value data.

Credentials & Honors:

  • Master of Science – Physics
  • Licensed Professional Engineer
  • Graduate of DOD Program Management Course
  • Physics honor society – Sigma Pi Sigma (SPS)
  • Academic honors – Phi Kappa Phi (FKF)
  • PMI Metrics SIG Scholar Award (2007)
  • PMI Eric Jenett Award (2007)
  • EVM Europe Award (2013)
  • CPM Driessnack Award (2014)
  • Australian Project Governance and Control Excellence Symposium established the annual Walt Lipke Project Governance and Control Award (2017)

Mr. Lipke can be contacted at [email protected]