CORRELATION ANALYSIS OF NAVY FLIGHT MISHAPS
Loading...
Authors
Kang, Ho Kyung
Subjects
Navy Aviation Mishap
prediction
supervised learning models
Random Forests
Multinomial Logistic Regression
flight mishap
mishap class
prediction
supervised learning models
Random Forests
Multinomial Logistic Regression
flight mishap
mishap class
Advisors
Yoshida, Ruriko
Date of Issue
2023-06
Date
Publisher
Monterey, CA; Naval Postgraduate School
Language
Abstract
Flight mishaps are a crucial part of military aviation communities’ operations because they negatively impact our military. The U.S. Navy’s Naval Safety Center Annual Reports document mishap matters to include total mishap numbers each year, and aircraft model. The reports display whether the current year’s mishaps have increased or decreased over the previous five years, but fail to provide any further analysis regarding correlations related to mishaps. The purpose of this research is to conduct a correlation analysis on the classes of flight mishaps through supervised learning techniques (i.e., Random Forest and Multinomial Logistic Regression analyses) to identify highly correlated variables related to mishap Classes A, B, and C. Naval Safety Command provided 10 years of flight mishap reporting data, which contained 754 observations with 13 variables. We found the “mishap type” and “aircraft model” variables correlate to the flight mishap classes the most, when predicting flight mishap classes. Mishap type is used to standardize mishap reporting and data collection; based on the Naval Aviation Safety Management System, there are 21 mishap types that may be selected during the mishap investigation process. Aircraft model describes the specific aircraft platform(s) that is involved in a mishap, including 26 aircraft models. Our correlation analysis provides the first step towards finding out the root causes of the flight mishaps and offers insights for future researchers.
Type
Thesis
Description
Series/Report No
Department
Operations Research (OR)
Organization
Identifiers
NPS Report Number
Sponsors
Funding
Format
Citation
Distribution Statement
Approved for public release. Distribution is unlimited.
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.