Can the Air Force Solve Its Spares Forecasting Problem

Abstract

Since the early 1980's spending by the U.S. Air Force for reparable for peacetime use has exceeded $2.0 billion per year. In the same period, Air Force estimates of future spares requirements have come under greater critical scrutiny, both within outside the Air Force. The Air Force Logistics Command (AFLC) has developed a regression-based forecasting model, ALERT (Air Logistics Early Requirements Technique), to improve POM (Program Objective Memorandum) estimates of future funding requirements for spares. ALERT predicts the output of the Air Force's budgeting and execution system for spares, the 'D041' system. To make its predictions, ALERT relies heavily on early D041 estimates, supplemented by age and value-of-the-fleet data by weapon system. Volatility in the underlying D041 system, in the form of fluctuating estimates for the same year, prevents ALERT from being able to make stable, accurate forecasts for POMs. ALERT also has conceptual problems, and its precision at the weapon-system level is poor. The conclusion is that ALERT will not solve the Air Force's credibility problem regarding spares. To solve the problem, the requirements system should be used to track and control requirements for spares-not just compute them.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1986
Accession Number
ADA176031

Entities

People

  • Christopher H. Hanks

Organizations

  • LMI

Tags

Communities of Interest

  • Air Platforms
  • C4I
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Air Force
  • Air Force Budgets
  • Air Force Facilities
  • Aircrafts
  • Availability
  • Budget Estimates
  • Classification
  • Computer Programming
  • Databases
  • Department Of Defense
  • Logistics
  • Logistics Management
  • Materials
  • Precision
  • Procurement
  • Weapon Systems

Readers

  • Aerospace logistics and air mobility.
  • Computational Modeling and Simulation
  • Logistics and Supply Chain Management.