Author Information

Dean SutherlandFollow

Is this project an undergraduate, graduate, or faculty project?

Undergraduate

Project Type

individual

Campus

Worldwide

Authors' Class Standing

Dean Sutherland, Senior

Lead Presenter's Name

Dean Sutherland

Lead Presenter's College

DB College of Business

Faculty Mentor Name

Burak Cankaya

Abstract

This study aims to analyze aircraft incident data from the Federal Aviation Administration (FAA) between 1978 and to current for the top five largest airlines in the United States. The goal is to predict conditions that have a higher chance of causing injury and to determine the best machine-learning model for this prediction. The target variable of "injuries occurred" is processed through Machine Learning methods such as Deep Learning, Logistic Regression, and Support Vector Machine. While analyzing and understanding the incidents, the text mining methodology is used in real-world incident remarks reports. The main scope of the study is not only to predict the injuries but also to understand and explain the complex fault mechanism that creates these incidents and recommend actions for these airlines to reduce the number and severity of these incidents by combining inferences between variables with Subject Matter Expertise in the area. This study provides insights into improving safety in the aviation industry, commonalities of injuries, and how it contributes to the literature by predicting injuries in legacy airlines, explaining the patterns that create incidents, and creating prescriptions to reduce injuries that happen on airlines by using explainable Machine Learning models.

Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?

No

Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?

Yes, Spark Grant

Share

COinS
 

Understanding the Incidents on Legacy Airlines with Machine Learning: Case Study Top 5 US Airlines

This study aims to analyze aircraft incident data from the Federal Aviation Administration (FAA) between 1978 and to current for the top five largest airlines in the United States. The goal is to predict conditions that have a higher chance of causing injury and to determine the best machine-learning model for this prediction. The target variable of "injuries occurred" is processed through Machine Learning methods such as Deep Learning, Logistic Regression, and Support Vector Machine. While analyzing and understanding the incidents, the text mining methodology is used in real-world incident remarks reports. The main scope of the study is not only to predict the injuries but also to understand and explain the complex fault mechanism that creates these incidents and recommend actions for these airlines to reduce the number and severity of these incidents by combining inferences between variables with Subject Matter Expertise in the area. This study provides insights into improving safety in the aviation industry, commonalities of injuries, and how it contributes to the literature by predicting injuries in legacy airlines, explaining the patterns that create incidents, and creating prescriptions to reduce injuries that happen on airlines by using explainable Machine Learning models.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.