民航预先飞行计划依据中新开航线客流预测模型构建  

Construction of passenger flow prediction model for new routes in civil aviation advance flight planning basis

作　　者：刘祥彬 Liu XiangBin(Middle South Regional ATMB.CAAC,Guangzhou Guangdong 510000)

机构地区：[1]中国民用航空空中交通管理局,广东省广州市510000

出　　处：《现代科学仪器》2024年第4期232-238,共7页Modern Scientific Instruments

摘　　要：随着经济发展,为了满足大众的需求,航空公司开辟了全新的航线。然而现有的预测方法无法高要求地满足这些新航线的客流量判断。鉴于此,研究首先对客流量影响因素进行分析和总结,随后选取了支持向量回归(Support Vector Regression,SVR)算法作为基础,引入了高斯径向基(Radial Basis function,RBF)核心函数进行优化。实验结果表明,SVR-RBF模型的客流量预测准确率最高为89.7%、偏差低于53.4%,与同类型的客流量预测模型相比,SVR-RBF模型的预测值与真实客流量值相差较小。综上所述,SVR-RBF模型能更好地预测新开航线的客流量,能够帮助航空公司满足大众市场的需求,为民航事业的发展提供了有效的理论支持。With the development of economy,airlines have opened up brand new routes to satisfy the public demand.However,the existing forecasting methods can not meet the high requirements of these new routes to determine the passenger flow.In view of this,the study firstly analyzes and summarizes the influencing factors of passenger flow,and then selects the Support Vector Regression(SVR)algorithm as the basis,and introduces the Gaussian Radial Basis function(RBF)core function for optimization.The experimental results show that the SVR-RBF model has the highest passenger flow prediction accuracy of 89.7%,the deviation is lower than 53.4%,and the difference between the predicted value and the real passenger flow value of the SVR-RBF model is small compared with the same type of passenger flow prediction model.In summary,SVR-RBF model can better predict the passenger flow of new routes,can help airlines to meet the demand of the mass market,and provides an effective theoretical support for the development of civil aviation.

关 键 词：新开航线 客流预测 支持向量回归 高斯径向基 核心函数 

分 类 号：F560.5[经济管理—产业经济]