Using Predictive Analytics to Detect Major Problems in Department of Defense Acquisition Programs
Abstract
This research provides program analysts and Department of Defense (DoD) leadership with an approach to identify problems in real-time for acquisition contracts. Specifically, we develop optimization algorithms to detect unusual changes in acquisition programs' Earned Value data streams. The research is focused on three questions. First, can we predict the contractor provided estimate at complete (EAC)? Second, can we use those predictions to develop an algorithm to determine if a problem will occur in an acquisition program or sub-program? Lastly, can we provide the probability of a problem occurring within a given timeframe? We find three of our models establish statistical significance predicting the EAC. Our four-month model predicts the EAC, on average, within 3.1 percent and our five and six-month models predict the EAC within 3.7 and 4.1 percent. The four-month model proves to present the best predictions for determining the probability of a problem. Our algorithms identify 70% percent of the problems within our dataset, while more than doubling the probability of a problem occurrence compared to current tools in the cost community. Though program managers can use this information to aid analysis, the information we provide should serve as a tool and not a replacement for in-depth analysis of their programs
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2012
- Accession Number
- ADA557925
Entities
People
- Austin W. Dowling
Organizations
- Air Force Institute of Technology