Forecast Error Metrics for Navy Inventory Management Performance

Abstract

This research establishes metrics for determining overall Navy secondary inventory forecasting accuracy when compared to actual demands at the Naval Inventory Control Point (NAVICP). Specifically, two performance metrics are introduced: the average performance index (API) and the median absolute deviation performance index (MPI). API measures forecasting accuracy of secondary inventory when compared against demand or forecast performance over a four-quarter period. MPI measures the quarterly variability of forecast errors over the same period. The API and MPI metrics allow for the identification of poorly forecasted NAVICP secondary inventory items. The metrics can be applied to entire inventories or subsets of items based on type, demand, or cost. In addition, the API metric can be used to show overall inventory performance, providing NAVICP with a graphical means to assess forecasting performance improvements (or degradations) over time. The new forecasting accuracy methods developed in this research will allow the Navy to continually gauge the overall health of their inventory management practices and provide a method for improving forecasting accuracy. Additionally, they will assist NAVICP in complying with DoD directives that require NAVICP to monitor and continually develop improvements to inventory management practices.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2011
Accession Number
ADA543431

Entities

People

  • Kenneth J. Jackson

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Human Systems

DTIC Thesaurus Topics

  • Accuracy
  • Data Sets
  • Delphi Method
  • Department Of Defense
  • Directives
  • Errors
  • Governments
  • Identification
  • Inventory
  • Inventory Control
  • Lead Time
  • Naval Aviation
  • Operations Research
  • Statistical Analysis
  • United States
  • United States Government
  • Uss Ronald Reagan

Readers

  • Computational Modeling and Simulation
  • Logistics and Supply Chain Management.
  • Parallel and Distributed Computing.