民航客运量预测方法研究综述  

Research Overview of Civil Aviation Passenger Traffic Forecasting Methods

作　　者：徐海文[1] 令海龙 夏思薇 XU Haiwen;LING Hailong;XIA Siwei(School of Science,Civil Aviation Flight University of China,Guanghan 618307,Sichuan,China;Air Traffic Management School,Civil Aviation Flight University of China,Guanghan 618307,Sichuan,China)

机构地区：[1]中国民用航空飞行学院理学院,四川广汉618307 [2]中国民用航空飞行学院空中交通管理学院,四川广汉618307

出　　处：《科技和产业》2024年第13期133-143,共11页Science Technology and Industry

基　　金：中央高校基本科研业务费专项资金(PHD2023-054)。

摘　　要：为了提升民航运行效率、准确预测客运量,促进其可持续发展,采用分类学方法将民航客运量预测方法划分为传统统计学、机器学习、组合模型3大类。详述各类方法的改进原理、效果和应用,通过数据处理、权重调整、参数优化和结构改进提高准确性,并总结组合模型相对单一模型的优势。实证研究结果表明,组合预测模型相较于单一模型具有更高的准确性,并指出结合人工智能和大数据技术的发展趋势,构建优秀的组合预测模型将是提高准确性的潜在研究方向。In order to improve the operating efficiency of civil aviation,accurately predict passenger traffic and promote the sustainable development of civil aviation,a taxonomy is used to classify civil aviation passenger traffic forecasting methods into traditional statistical methods,machine learning and combination models.The improvement principle,effect and application of each method are introduced in detail,the precision is improved by data processing,weight adjustment,parameter optimisation and structure improvement,and the advantages of combinatorial model over single model are summarized.Empirical results show that prediction model have higher accuracy than single models,and it is pointed out that combining the development trend of artificial intelligence and big data technology,constructing excellent combinatorial prediction models is a potential research direction to improve prediction accuracy.

关 键 词：民航客运量 时间序列预测 机器学习 神经网络 组合预测 

分 类 号：U8[交通运输工程]