Please use this identifier to cite or link to this item:
https://dair.nps.edu/handle/123456789/1760
Title: | Quantifying the Year-by-Year Cost Uncertainty of Major Defense Programs |
Authors: | David M. Tate Michael R. Guggisberg |
Keywords: | Year-By-Year Cost Uncertainty Cost Risk Functional Regression Weibull Curves Functional Principal Component Analysis FPCA |
Issue Date: | 13-May-2019 |
Publisher: | Acquisition Research Program |
Citation: | Published--Unlimited Distribution |
Series/Report no.: | Acquisition Management SYM-AM-19-070 |
Abstract: | To a first approximation, acquisition programs never spend what they originally said they would spend when they began. In fact, the uncertainty in initial funding profile estimates is much larger than is generally understood; the possibility of program cancellations, restructurings, truncations, and block upgrades are often not accounted for. Worse yet, all of this uncertainty arises in a context in which programs must fit within annual budgets it is not enough to only spend as much as you said you would; you must also spend it when you said you would, or problems ensue. In 2018, we presented a methodology that uses historical program outcomes to characterize the year-by-year development and procurement cost risk associated with a major acquisition program. That work used functional regression to characterize changes in development profiles, modeled as Weibull curves. This paper improves and extends that work, using a novel application of Functional Principal Component Analysis (FPCA) to characterize the distributions of future RDT&E and Procurement profiles of both new and continuing acquisition programs. |
Description: | Acquisition Management / Defense Acquisition Community Contributor |
URI: | https://dair.nps.edu/handle/123456789/1760 |
Appears in Collections: | Annual Acquisition Research Symposium Proceedings & Presentations |
Files in This Item:
File | Size | Format | |
---|---|---|---|
SYM-AM-19-070.pdf | 592.12 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.