Graduation Year

2024

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Yu Zhang, Ph.D.

Committee Member

Fred Mannering, Ph.D.

Committee Member

Sameer Alam, Ph.D.

Committee Member

Joseph Post, Ph.D.

Committee Member

Peng Chen, Ph.D.

Keywords

Collaborative Decision Making, Machine Learning, Traffic Management

Abstract

Extensive research in predicting annual passenger throughput at airports has been conducted, aiming at providing decision support for airport construction, aircraft procurement, resource management, flight scheduling, etc. However, few studies have been conducted on how airport operational throughput is affected by convective weather in the vicinity of the airport and how to predict short-term airport operational throughput. Convective weather near the airport could make arrivals miss their positions in the arrival stream and reduce airfield efficiency in terms of utilizing runway capacities. This research leverages the learning-based method (MB-ResNet model) to predict airport hourly throughput. It uses the Hartsfield–Jackson Atlanta International Airport (ATL) as the case study to demonstrate the developed method. This research uses the Rapid Refresh model (RAP) data from the National Oceanic and Atmospheric Administration (NOAA) to indicate convective weather. Although RAP data, along with its predecessor RUC, has been extensively applied in air traffic management (ATM) research, particularly for obtaining wind and pressure data on a large scale to support trajectory modeling. In this study, we developed a novel approach to leverage RAP data specifically for predicting convective weather patterns. This research aims to explain variations in airfield efficiency. By carefully decoding, cleaning, and pre-processing RAP data, we demonstrated its potential to enhance predictions related to airfield operations. Applying machine learning and deep learning in air traffic management remains a promising field for aviation researchers, as these advanced artificial intelligence techniques can harness large-scale aviation data to improve the predictability of the national airspace system and, consequently, operational efficiency. The short-term airport operational throughput forecasted in this study provides valuable insights for air traffic controllers and airport managers, facilitating better resource allocation and operational improvements.

Download

Share

COinS