Predicting FA-18 Squadron Readiness and Quarterly Flight Hour Execution Using Machine Learning
Abstract
Given manning-training-equipment datasets from Naval FA-18 squadrons, a machine learning model for determining the monthly mean number of mission capable jets per squadron is created. This model is then extended and used as an input to create an ensemble of models determining the flight hour execution of a squadron over a three-month period. The ensemble of models is then used to predict squadron performance and readiness, and can correctly classify a squadron's future performance with 75% accuracy 90-days in advance.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2019
- Accession Number
- AD1085789
Entities
People
- Benjamin Michlin
- Charles Yetman
- Dean Lee
- Josh Duclos
- Rick Cruz
- Ruey Chang
- Vincent Siu
Organizations
- Naval Information Warfare Center Pacific