Posted by SE/IT/PM Trend Analysis
Methods Track
Downloadable Files:
Abstract:
Systems Engineering (SE), Integration and Test (I&T), and Program Management (PM) costs have traditionally been difficult items to estimate at any given point in a program. Cost Estimating Relationships (CERs) are often used in practice to predict the ultimate SE/IT/PM cost. The utility of a CER naturally lessens as a program matures since CERs do not typically take into account known performance to date or specific details of the program that are not captured in the CER variables. Many people view these functions as a level of effort and estimate them as a function of time, which is dependent on predicted schedule, but does not account for program performance to date. Some widely-used methods that do take the actual performance into account include earned-value, staffing- level methods, and fixed-percentage extrapolation. Each of these methods has weaknesses that can be avoided using the presented technique.
This paper presents an alternative to existing methods, using both historical data on similar programs and actual performance on the current program to project final costs. It builds on concepts first applied by an NRO contractor in 2004 for a space system estimate at complete. This method examines the historical trend of cumulative SE/IT/PM versus percentage of time complete; considered the “bathtub effect.” Preliminary analysis demonstrated a consistent shape of the curve confirmed by several historical programs. Variations on this concept are explored, as well as possible applications of risk.
Estimating relationships are shown for multiple levels of SE/IT/PM, including System of Systems, Space Segment and Ground Segment. It is important to note that while the study was performed using space system data, the method could easily be applied to any Department of Defense or Intelligence Community program cost estimate. Challenges for future research are also identified, most prominently the need for more time-phased data. With the addition of future data, it appears this method can offer the potential for more realistic estimates of SE/IT/PM costs given known information to date.
Author(s):
Alex Wekluk
Alex Wekluk has been a cost analyst for Northrop Grumman since 2005. Mr. Wekluk has supported several Intelligence Community customers, performing model development, method improvement, and lifecycle cost estimating. Previous work experience includes several years with the Naval Surface Warfare Center doing mechanical design. Mr. Wekluk has a Bachelor of Science in Mechanical Engineering from the Virginia Polytechnic Institute and State University. Certifications include SCEA CCE/A and the State of Virginia EIT for passing the Fundamentals of Engineering exam.
Nathan Menton
Nathan Menton is a Junior Cost Analyst who joined MCR Federal in 2005. With MCR he assisted in cost model development and cost estimating at the NRO Cost Group for a year and a half and is currently working in MCR’s East Coast Space Division. Mr. Menton earned a B.S. in Statistics with a minor in Mathematics from The George Washington University. He will be attending George Mason University in the fall to begin graduate degree work in the Statistical Science.