Posted by

SE/IT/PM Estimating Approaches for Service-Oriented Architecture Environments

Software/Hardware/Tools Track

Downloadable Files:



Within the Intelligence Community, there is a movement towards breaking down stovepiped architectures and implementing service-oriented architectures (SOAs). The idea is that new programs will be integrated into an existing SOA framework, thus allowing data sharing between systems. Estimating costs, particularly Systems Engineering/Integration and Test/Program Management (SE/IT/PM) costs, for these new programs that will be sharing the system poses several problems. The questions that arise from this are as follows:

1) The overwhelming majority of data that is available is the cost data for the development of the entire framework. Using regression analysis, an analyst can get a reasonably good CER and y-intercept for the SE/IT/PM cost of the whole framework based on the prime mission equipment (PME), or hardware and software development and engineering. The argument can be made that the coefficient of the regression can be applied to the PME associated with a new program being integrated into the larger system; however, if the framework is not fully mature it may require alterations to accept new programs. Because integrating into the immature framework involves more SE/IT/PM than just integrating the new hardware and software involved, how can this analysis be applied to a new program “plugging into” the shared framework?

2) Using the above-mentioned CER approach, the SE/IT/PM for a program rides along with PME. However, SE/IT/PM often begins before spending on PME and ends after spending on PME has completed. How can the SE/IT/PM costs be most accurately phased using data from historical data collection efforts?

This paper will explore these questions and will discuss several possible options for estimating SE/IT/PM in the development period.


Hilary G. Eckberg
Hilary G. Eckberg is an Operations Research analyst for Northrop Grumman IT, Intelligence Group (TASC), where she supports the intelligence community. Her activities include cost estimation and analysis and cost data collection and normalization. She also participates on ICE teams conducting internal cost and risk assessments. Hilary received her Bachelor of Fine Arts in Technical Theatre/Scenery Construction from Boston University in May 2001 and her Masters of Science in Computational Operations Research from The College of William and Mary in December 2006.

Kevin J. Snyder
Kevin J. Snyder is an Operations Research analyst for Northrop Grumman IT, Intelligence (TASC), where he provides cost analysis and estimating support to a variety of agencies throughout the Intelligence Community. Prior to his work in the Intelligence Community, he worked on the TASC cost team for the US Navy’s DD(X) destroyer program. Kevin received his bachelors of science from The College of William & Mary in Mathematics in May 2004. He received his Masters of Science in Operations Research from George Mason University in December 2007.