David S. Christensen

Delivering projects on time

OVERVIEW: Meaningful risk analysis can dramatically improve the likelihood of project managers delivering major system development projects on time. At issue are the questions, “Which risk events are most likely to occur?” and “Which risk events will have the most severe impact on the project schedule?” A study of 22 major aerospace system development programs produced a prompt list to support project managers engaged in risk identification. The study also yielded a risk map based upon an empirical study of the reasons for adverse schedule variances on the 22 programs over a ten-year period. The lack of required parts and materials when needed, a lack of requisite information, design changes, and difficulty starting, are among the risk events that most warrant managerial attention and appropriate risk mitigation and control.

Is the CPI-Based EAC a Lower Bound to the Final Cost of Post A-12 Contracts?

Abstract Christensen (1994b, 1999) describes several methods to evaluate the predicted final cost of a defense acquisition contract, termed the “Estimate at Completion” (EAC). One of the methods uses the EAC derived from the cumulative Cost Performance Index (CPI) as a lower bound to the final cost of a defense contract. The method was derived from Department of Defense (DOD) experience on hundreds of defense programs completed in the 1970s and 1980s. This study tests the validity of the rule on contracts completed after 1991. Based on an analysis of 52 defense acquisition contracts completed after 1991, the rule is still valid for CPI-based EACs computed in the early and middle stages of a contract, but not valid in the late stage.

Is the Cumulative SCI-based EAC an Upper Bound to the Final Cost of Post-A12 Defense Contracts?

Abstract Christensen (1999) describes several methods to evaluate the predicted final cost of a defense acquisition contract, termed the “Estimate at Completion” (EAC). One of the methods uses the EAC derived from the cumulative Schedule-Cost Index (SCI) as a potential upper bound to the final cost of a defense contract. The method was derived from Department of Defense (DoD) experience on hundreds of defense contracts completed in the 1970s and 1980s, and used by Beach (1990) to evaluate the reasonableness of contractor and government estimates of the final cost of the A12 program. This study tests the validity of the rule on 120 contracts completed after the A-12 cancellation in 1991. Results show that the average cumulative SCI-based EAC is significantly lower than the average final cost of defense contracts in the early and middle stages of completion. Using it as a lower bound may result in more realistic estimates of final cost.

A Note on Bias in DOD Inflation Forecasts

Abstract The importance of accurate estimates of inflation to the DOD is enormous. The end of the cold war marked the beginning of a “cost war” for the DOD. According to the GAO (1994, 1996), pressure to cut billions from the defense budget may encourage overly optimistic estimates of inflation that could lead to budgetary shortfalls, funding instability, and reduced military capabilities. The GAO and others have asserted that DOD inflation estimates are downwardly biased. We test this assertion by examining historical DOD forecasts against actual inflation as measured by the Gross National Product Implicit Price Deflator from 1979 to 1996. Based on the Wilcoxon signed-ranks test, results indicate significant upward bias in near-term projections, and no bias in projections spanning 2–5 years.