A Review of Cost Performance Index Stability

Abstract

This study examines approaches currently used to determine when Cost Performance Index (CPI) stability occurs. The CPI indicates the cost performance efficiency of the work the contractor has accomplished to date; however it has value as a predictor of future contractor cost performance given that the CPI can be declared stable. Knowing that the CPI is stable allows government personnel to project the final cost of the contract and, if a cost overrun is projected, to determine the likelihood that the contractor can recover. The range method was used to test for stability of cumulative and non-cumulative (three month, six month, and six month moving average) CPI values. The method measured the range of the CPI values that occurred after the 50 percent contract completion point. The results (range method sensitivity analysis) indicated that the cumulative CPI is more stable than the other CPIs examined, stabilizing from the 20 percent completion point. As a caveat however, the method of least squares showed that the cumulative CPI does tend to decline within the allowable range limit as the contract proceeds. Thus the final cumulative CPI value can be expected to be lower than the cumulative CPI value observed at the 20 percent contract completion point.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1991
Accession Number
ADA246621

Entities

People

  • Scott R. Heise

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Human Systems
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Aircrafts
  • Airframes
  • Anti-Tank Missiles
  • Artificial Satellites
  • Control Systems
  • Cost Overruns
  • Databases
  • Department Of Defense
  • Fighter Aircraft
  • Fixed Price Contracts
  • Helicopters
  • Measurement
  • Navy
  • Rockets
  • Statistical Analysis
  • Surface To Air Missiles
  • Test Equipment

Readers

  • Industrial Economics
  • Mathematics or Statistics
  • Systems Analysis and Design