Author

Joshua Mote

Date of Award

2025

Document Type

Thesis

Degree Name

Master of Science in Operations Research (MSOR)

Department

Industrial and Systems Engineering and Engineering Management

Committee Chair

Howard Chen

Committee Member

Sean Freeman

Committee Member

Bryan Mesmer

Research Advisor

Howard Chen

Subject(s)

Meteorology in aeronautics, Weather forecasting, Machine learning, Neural networks (Computer science)

Abstract

Weather forecasting is critical to minimize risk and maximize efficiency for flying operations and is challenging due to the uncertainty involved in atmospheric changes. Operational area weather is becoming more critical with the growing reliance on air domain for transportation. This study investigates whether the forecasting accuracy using a Hybrid Long-Short Term Memory (LSTM) Artificial Neural Network (ANN) outperforms the San Antonio International Airport (KSAT) Terminal Aerodrome Forecast (TAF) by integrating diverse data sources – regional Meteorological Aerodrome Report (METAR), TAF, Avian Advisory System (AHAS) Data and geomagnetic activity K Index Data. When tested on one year of historical data, the Hybrid LSTM model exceeded the KSAT TAF’s average mean absolute error (MAE) for eleven continuous weather parameters by 64% and exceeded the accuracy for nine binary weather parameters by an average of 4%, highlighting the performance of the ANN compared to the legal forecast for KSAT.

Share

COinS
 
 

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.