Using ANOVA and Multinomial Logistic Regression to Analyze Defense Acquisition Executive Summary (DAES) and Acquisition Program Baseline (APB) Milestone Estimates to Determine Contributing Factors to Schedule Slips

Abstract

The process of acquiring Major Defense Acquisition Programs (MDAPs) consumes one of every eight dollars in the Department of Defense budget, costing the Department nearly $90 billion in FY2019. Due to their size and their importance to the warfighter, the success of these programs is paramount to U.S. military dominance for the decades ahead. However, these programs often encounter difficulties, resulting in schedule slips beyond original estimates. During 2014, Under Secretary of Defense for Acquisition, Technology, and Logistics Kendall, instituted the Better Buying Power initiative, which aimed to address the problem of cycle-times and schedule slips. By using analysis of variance (ANOVA) and multinomial logistic regression models to analyze Milestone Current Estimates from Defense Acquisition Executive Summaries (DAES) reports, as well as Milestone Threshold Dates from acquisition program baseline (APB) reports, we seek to determine which factors lead to higher likelihoods of schedule slips. Determining these root causes could inform future acquisition decisions and lead to better original estimates. Our research found that significant mean rank differences exist in all categories and across all schedule slip variables and that as one-month increases occur in the four derive schedule slip variables, there are significant odds that various factor categories are likely to have contributed to that increase in schedule slip.