Date of Award
Master of Science
Department of Mathematics and Statistics
Edward D. White, PhD
Cost Growth in Department of Defense (DoD) major weapon systems has been an on-going problem for more than 30 years. Previous research has demonstrated the use of a two-step logistic and multiple regression methodology to predicting cost growth produces desirable results versus traditional single-step regression. This research effort validates, and further explores the use of a two-step procedure for assessing DoD major weapon system cost growth using historical data, We compile programmatic data from the Selected Acquisition Reports (SARs) between 1990 and 2001 for programs covering all defense departments. Our analysis concentrates on cost growth in the research and development dollar accounts for the Engineering and Manufacturing Development phase of acquisition. We investigate the use of logistic regression in cost growth analysis to predict whether or not cost growth will occur in a program. If applicable, the multiple regression step is implemented to predict how much cost growth will occur. Our study focuses on four of the seven SAR cost growth categories within the research and development accounts - schedule, estimating, support, and other. We study each of these four categories individually for significant cost growth characteristics and develop predictive models for each.
DTIC Accession Number
Bielecki, John V., "Estimating Engineering and Manufacturing Development Cost Risk Using Logistic and Multiple Regression" (2003). Theses and Dissertations. 4182.