航班着陆调度的多目标优化研究

摘要

航班着陆调度问题是多目标优化问题，难以使用最优化方法求解。为了解决这一难题，以减少航班延迟时间和降低飞行延误成本为目标，提出一种整合的启发式方法。该方法使用吱呀轮算法 SWO（Squeaky-Wheel Optimization）进行导向式搜索，并利用改进的 GA 充分扩展 SWO 的搜索空间，最后通过合理整合 GA 和 SWO，取得求解效率和求解质量的提高。通过实验仿真对比表明该算法能高效求解该问题，满足了实时调度的需求，同时求解质量也优于其他启发式算法，节省了更多降落时间和成本。%Flight landing scheduling is a multi-objective optimisation problem,which is difficult to be solved with optimisation method.To deal with this difficulty,we propose an integrated heuristic algorithm targeted at cutting down the flight delay time and reducing flying delay cost.The method uses Squeaky-Wheel Optimisation (SWO)to carry out the guided search,and uses improved genetic algorithm (GA)to sufficiently expand the search space of SWO,at last by reasonable integrating the GA and SWO,it achieves the improvement in both efficiency and quality of solution.It is demonstrated through experimental simulation that the algorithm can solve the problem efficiently and meet the requirement of real-time scheduling.At the same time,its solution quality is also better than other heuristic algorithms,and saves more landing time and cost.