Considerations for Obsolescence Prediction
Abstract
Our research considers statistical and mathematical techniques that demonstrate potential for improving the predictive capabilities used by the Navy in modeling hardware obsolescence. In Section 1, we define the type of obsolescence considered in our research (Section 1.1) and discuss common predictive methods in the obsolescence literature (Section 1.2). In Section 3, our experiments comparing the performances of random forests and random survival forests for modeling data with right-censored responses are described.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 29, 2019
- Accession Number
- AD1092377
Entities
People
- Anthony Leclerc
- Emily Nystrom
Organizations
- Naval Information Warfare Center Atlantic