Forecasting of Secondary Item Returns.
Abstract
A family of Return Rate models using a Kalman Filter to estimate unserviceable returns per unit of program (Flying Hours or Item Density) is examined. A comparison--based on several measures of forecast error--is made with the current forecasting model of unserviceable returns per unit of demand. The current model produces forecast errors 10 to 20 percent higher than the Kalman Filter models.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1976
- Accession Number
- ADA030343
Entities
People
- Richard M. A. Urbach