Optimizing aircraft departure and arrival sequencing using genetic algorithms

• Main Author: Ng, Timothy Jefferson Wei Han
• Other Authors: Khoo Li Pheng
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering; DRNTU::Engineering::Aeronautical engineering::Air navigation; DRNTU::Engineering::Aeronautical engineering::Accidents and air safety; DRNTU::Business::Management::Mathematical models; DRNTU::Engineering::Aeronautical engineering::Aviation
• Online Access: http://hdl.handle.net/10356/61303
• Institution: Nanyang Technological University

With a major increase in air transport projected over the next few decades, there is an increasing need for airports to fully utilize their throughput by minimizing the time required for a given set of aircraft to land on a runway. The aim of this study is to develop a novel algorithm to optimize the sequence of aircraft departing and arriving at Changi Airport Terminal 2 using an evolutionary algorithm known as Genetic Algorithms (GA). After reviewing past work on the Aircraft Landing Problem to understand the real world constraints the new algorithm is developed, integrating important concepts such as departing aircraft, maximum delay, and earliest possible arrival time. This is done by introducing an original reproduction operator and objective function. Subsequently a TABU search function is incorporated into the GA to enhance its capabilities. The GA is also modified to perform dynamic optimizations for newly arrived aircraft using the concept of Receding Horizon Control (RHC). An analysis of the results shows that the static GA is able to find the optimum solution for the 20 aircraft scenario quickly due to position shift constraint. The addition of the TABU function was found to not be able to improve results significantly due to the fact that multiple solutions with equally good results exist. Finally, the 2 forms of dynamic GA developed were both functional. However, each traded run-to-run stability for better results and vice versa.