Estimating the final cost of a DoD acquisition contract
| Published date | 01 March 2011 |
| Pages | 190-205 |
| DOI | https://doi.org/10.1108/JOPP-11-02-2011-B002 |
| Date | 01 March 2011 |
| Author | Steven P. Tracy,Edward D. White |
| Subject Matter | Public policy & environmental management,Politics,Public adminstration & management,Government,Economics,Public Finance/economics,Texation/public revenue |
JOURNAL OF PUBLIC PROCUREMENT, VOLUME 11, ISSUE 2, 190-205 SUMMER 2011
ESTIMATING THE FINAL COST OF A DOD ACQUISITION CONTRACT
Steven P. Tracy and Edward D. White*
ABSTRACT. The most common technique to determine the predicted final
cost of a Department of Defense (DoD) acquisition contract, or the Estimate
at Completion (EAC), involves the use of performance indices to adjust the
EAC. Other methods including simple linear regression and time series
analysis have been developed to predict the final cost, but these methods
are not widely publicized or have limited applicability. As a potential remedy,
this research utilizes the historical contract data reported in the Defense
Acquisition Executive Summary database and provides to the analyst a set of
five working multiple regression models. Useful over the life of the contract,
they accurately predict the final cost of the average major weapons system
contract using contractor Cost Performance Report data.
OVERVIEW
Earned value management system (EVMS) in the U.S. government
acquisition process provides a means of organization for project
schedule, budget, and planning components that can produce
forecasts and status determinations. EVMS equips program
managers with the capability to forecast results used in the decision-
making process. The measures highlighted by EVMS methodology
provide the inputs for these forecasts, termed Estimates at
Completion (EACs). The EAC is the estimated final cost of a
program’s contract value. The EAC is typically computed by using the
----------------------
* Steven Tracy, MS, is a weapon systems cost analyst supporting Naval Air
Systems Command, Naval Air Station Patuxent River, Maryland. Edward
White, Ph.D., is an Associate Professor, Department of Mathematics and
Statistics, Air Force Institute of Technology. His research interests include
cost analysis, regression modeling, design of experiments, growth curves,
and biostatistics.
Copyright © 2011 by PrAcademics Press
ESTIMATING THE FINAL COST OF A DOD ACQUISITION CONTRACT 191
Cost Performance Index (CPI) or the Schedule Performance Index
(SPI) as products of the Earned Value Management System. The CPI
measures the value of the work completed to the actual cost and
equates to the ratio of Budgeted Cost of Work Performed (BCWP) to
Actual Cost of Work Performed (ACWP), while SPI equates to BCWP to
Budgeted Cost of Work Scheduled (BCWS).
Both CPI and SPI measure contractor performance for a given
time period or from contract award to present. When reported as a
cumulative metric for a contract’s performance, they are referred to
as Cum CPI or Cum SPI. Whether dealing with CPI/SPI or Cum
CPI/Cum SPI, these measures are computed by comparing the
contractor’s actual cost and schedule performance against the
contractor’s planned performance. For example, if BCWP < BCWS,
then the program is behind schedule for the specified time period. If
Cum ACWP > Cum BCWP, then this indicates a cost overrun for the
program for the entire length of the program from inception to current
report period.
To compute the EAC, analysts typically use the CPI or SPI
individually, a multiplication of the two indices, or some weighted
combination of each. For example, the Schedule Cost Index (SCI)
equals SPI*CPI. The Composite Index (CI) equates to W1*SPI +
W2*CPI, where W1+W2 equals 1, i.e. 100% (W1 and W2 are
numerical weights applied to SPI and CPI, respectively). From a
choice of these four indices, CPI, SPI, SCI, and CI, the EAC can be
determined from EAC = ACWP + (BAC – BCWP)/Index, where BAC is
the Budget at Completion.
Because the EAC could vary for a contract given a particular index
used, researchers have conducted studies to determine which, if any,
index reflects the most accurate EAC. Initial comparison papers
showed that the CPI demonstrated the best predictive index (Busse,
1977; Karsch, 1976). A later study re-examined this and concluded
with the SCI as the best predictor (Terry and Vanderburgh, 1993).
And in another study, Nystrom (1995) concluded that a Composite
Index method was a more accurate and stable prediction tool. With
no common ground in terms of universal ‘best’ index, Christensen et
al. (1995) looked at a diverse collection of studies both published
and unpublished to ascertain if such a thing existed. He concluded
that the accuracy of index-based formulas depends on the type of
acquisition system and the stage and phase of the contract. In other
To continue reading
Request your trial