An Improved Genetic Algorithm-Based Traffic Scheduling Model for Airport Terminal Areas

Abstract

This paper takes the airport terminal area as the main research content and combines genetic algorithm with airport terminal area analysis theory to analyze and study the traffic scheduling in the airport terminal area. Based on the study of traditional traffic scheduling techniques and key techniques of genetic algorithms, this paper participates in the actual project of genetic algorithm-based traffic scheduling, analyzes the requirements of the project, focuses on the design and implementation of the traffic scheduling algorithm module in the genetic algorithm-based traffic scheduling system, and conducts further research on the pathfinding by constraints submodule. In this paper, the flight approach and departure sequencing problem and runway allocation problem are the main research objects. The dynamic optimal scheduling model of flight approach and departure is established by considering the interests and demands of airlines and airports, and a new scheduling algorithm is proposed. In this paper, a brief introduction to the airport terminal area is given, and the feasibility of the approach/departure optimal scheduling is introduced from the perspective of airlines with a long-range parallel two-runway airport as the research background. Secondly, through the analysis of the flight approach and departure process and the study of the approach and departure cooperative optimization strategy, a single-runway flight approach and departure traffic scheduling model under the joint sequencing strategy is established with the optimization objective of minimizing the total flight delay time, and the model is solved by using the sliding time window algorithm. Then, based on the single-runway scheduling model, a multirunway multiobjective flight optimal scheduling model is established with the objectives of minimizing total delay time, increasing runway throughput per unit time and fairness of flight delay time allocation, and a dynamic algorithm (STW-GA) combining sliding time window algorithm and dual-structured chromosome genetic algorithm is proposed to solve the model.