民航中小机场运行风险管控措施效能评估研究  

Research on the Evaluation of the Effectivenessof Operational Risk Management for Medium or Small Airports

作　　者：唐卫贞[1] 黄婷 黄洲升 TANG Wei-zhen;HUANG Ting;HUANG Zhou-sheng(Civil Aviation Flight University of China,Guanghan 618000,China)

机构地区：[1]中国民用航空飞行学院,四川广汉618000

出　　处：《航空计算技术》2024年第5期1-5,10,共6页Aeronautical Computing Technique

基　　金：中国民用航空局安全能力建设项目资助(MHAQ2022032);中国民用航空局教育人才类资金项目资助(MHAQ2022015)。

摘　　要：机场作为关键的交通枢纽和经济引擎,必须实施安全风险管理,以确保其安全运营,为此,提出了危险源管控措施效能评估模型。采集多个机场的危险源数据,对其进行量化分析;利用GloVe将管控措施文本处理成词向量,通过文本卷积神经网络(TextCNN)模型,评估出风险管控措施的效果。通过对比实验,确定危险源管控措施效能评估模型的优劣。结果表明基于TextCNN的效能评估模型的准确率达到74.36%,而基于TextRNN的模型为69.24%。因此,TextCNN更适合用于此类评估任务。本研究为机场管理部门提供了一个新的视角和方法,以更有效地评估和管理各种危险源,从而提升整个机场的安全性能。As key transportation hubs and economic engines,airports must implement security risk management to ensure safe operations.Therefore,this study proposes an evaluation model for the effectiveness of hazard control measures,which firstly collects hazard source data from multiple airports,quantitatively analyzes them,and then uses GloVe to process the text of control measures into word vectors,and uses the text convolutional neural network(TextCNN)model to realize text classification and evaluate the effect of risk control measures.Finally,the advantages and disadvantages of the efficacy evaluation model of hazard control measures were determined.The results show that the accuracy of the performance evaluation model based on TextCNN is 74.36%,while that of the model based on TextRNN is 69.24%.Therefore,TextCNN is more suitable for such evaluation tasks.This study provides a new perspective and approach for airport authorities to more effectively identify,and manage various sources of hazard,thereby improving the safety performance of the entire airport.

关 键 词：机场风险管控 卷积神经网络 危险源 效能评估 自然语言处理 效能评估 

分 类 号：V35[航空宇航科学与技术—人机与环境工程]