Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


48546.pdf (9.93 MB)

ETD Abstract Container

Abstract Header

Digital Modeling of CLT Airport Surface Traffic: A Neural-Queuing Approach for Airport Traffic Prediction

McKenna, Liam
http://orcid.org/0009-0001-4499-0259
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1746703363290598

Abstract Details

2025, MS, University of Cincinnati, Engineering and Applied Science: Aerospace Engineering.
As the air travel industry grows, modernization of Air Traffic Control procedures are imperative to enabling new technology while maintaining high operational efficacy and safety. Conventional air travel between large airports via private airlines remains a large and demanding sector of our National Airspace System. Efforts to reduce delays and increase the efficiency and throughput of airports are of interest to stakeholders within the National Airspace System such as airlines, passengers, and Air Traffic Control personnel. Due to the (often contradictory) interests of both safety and efficiency for air travel, improvement of the traffic system requires careful research. Initiatives like FAA NextGen motivate researchers to implement new operational tactics in the field of Air Traffic Management, aiming to improve the efficacy of the system through predictable and flexible Traffic Management Initiatives (TMIs). On the airport surface side, common topics for ATM researchers include capacity planning, runway configuration management and departure pushback metering. All of these research topics require prediction of airport behavior as a means to optimize and validate the approaches in question. Methods of prediction that are currently employed by ATM researchers range from very simple to highly complicated, and current works are still working to accurately and simplistically produce traffic flow estimation for validating such decision support tools. In this work, we propose a method to capture the complex, long-term traffic flow behavior of an airport’s surface using a combination of numerical discrete time queueing models and machine learning, which we apply to Charlotte-Douglas International Airport (CLT). Our approach addresses some shortcomings faced by other simplified airport performance models and model-free approaches by encoding airport states with an Artificial Neural Network (ANN). The approach detailed in this work also performs aggregate and mesoscopic estimation, removing the need for extremely detailed agent-based techniques like microsimulation. We validate the accuracy and stability of the proposed Neural-Queue framework by comparing simulated traffic and congestion metrics produced in an extended simulation with historical traffic data.
Abhinav Sinha, Ph.D. (Committee Chair)
Rajnikant Sharma, Ph.D. (Committee Member)
Manish Kumar, Ph.D. (Committee Member)
142 p.
Aerospace Engineering
Neural Queueing; Queueing Theory; Airport Surface Traffic; Machine Learning; Simulation; CLT Airport

Recommended Citations

Citations

  • McKenna, L. (2025). Digital Modeling of CLT Airport Surface Traffic: A Neural-Queuing Approach for Airport Traffic Prediction [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1746703363290598

    APA Style (7th edition)

  • McKenna, Liam. Digital Modeling of CLT Airport Surface Traffic: A Neural-Queuing Approach for Airport Traffic Prediction. 2025. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1746703363290598.

    MLA Style (8th edition)

  • McKenna, Liam. "Digital Modeling of CLT Airport Surface Traffic: A Neural-Queuing Approach for Airport Traffic Prediction." Master's thesis, University of Cincinnati, 2025. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1746703363290598

    Chicago Manual of Style (17th edition)

ucin1746703363290598
129
© 2025, some rights reserved.Creative Commons License Digital Modeling of CLT Airport Surface Traffic: A Neural-Queuing Approach for Airport Traffic Prediction by Liam McKenna is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by University of Cincinnati and OhioLINK.
Release 3.2.12