基于UML-HMM的飞行员操作可靠度推断模型  

Pilot Operation Reliability Inference Model Based on UML-HMM

作　　者：朱荟群 刘星[1] 孙有朝[1] ZHU Hui-qun;LIU Xing;SUN You-chao(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学民航学院,江苏南京20016

出　　处：《航空计算技术》2018年第4期16-19,23,共5页Aeronautical Computing Technique

基　　金：国家自然科学基金与中国民航基金联合项目资助(U1333119);国防基础科研计划项目资助(JCKY2013605B002);国防科工局技术基础科研项目资助(Z052013B003)

摘　　要：为解决飞行员在不同飞行阶段中的操作可靠度不确定性问题,提出基于UML-HMM的飞行员操作可靠度推断模型。结合飞行标准操作程序(SOP),基于统一建模语言(UML)建立飞行员在不同飞行阶段的工作流时序图模型。基于飞行员工作流时序图建立隐马尔科夫模型(HMM)。以B737-300/NG飞机飞行阶段为例,得出了飞行员操作可靠度变化曲线,识别出了可靠度较低的阶段和致因,实例证明与实际情况相符,为设计人员和飞行员提供了针对性的建议和参考。In order to solve the uncertainty of operational reliability of pilots in different flight phases,a pilot operation reliability inference model based on UML-HMM is proposed. In conjunction with the flight standard operating procedure（SOP）,a workflow timing diagram model for pilots at different flight phases is established based on the unified modeling language（UML）. A hidden Markov model（HMM） is built based on the pilot workflow timing diagram. Taking the flight phase of B737-300/NG as an example,the pilot operation reliability curve is obtained,and the stage and cause of low reliability are identified. The example proves that it is consistent with the actual situation and provides the designer and pilot with targeted advice and references.

关 键 词：飞行员 操作可靠度 工作流 隐马尔科夫模型 人为差错 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]