【作者】 于海田；

【导师】 冯之浚；

【作者基本信息】 上海大学 ， 机械制造及其自动化， 2011， 博士

【摘要】 民用航空作为国民经济和社会发展的重要行业,是一个高技术、高投入、高风险的行业。安全是民航永恒的主题,航空安全管理的能力和水平是影响和制约航空发展的基本要素。探讨符合中国特点的民航安全管理技术与方法,构建民航持续安全框架,对于提高民航安全水平、实现民航持续安全具有重要意义。立足于解决当前民航安全分析与管理中的关键和难点问题,本文的主要研究内容和创新点如下:(1)民航安全管理与民航持续安全技术与方法民航安全系统是一个复杂巨系统,影响民航安全因素多,逻辑关系复杂,增加了民航安全管理的难度。将系统工程理论和方法引入到民航安全系统研究中,实现风险识别、安全分析与评价、安全预测和控制于一体的系统安全管理,能够发现民航安全系统中存在的潜在危险,提高安全管理的效益和效率。以实现民航长效安全运行机制为目的,构建了民航持续安全管理框架,研究了包括民航系统安全管理、过程安全管理、细节安全管理、信息安全管理和民航安全文化管理于一体的集成安全管理技术与方法。(2)民航维修系统差错分析技术民航维修工作是当前影响维修安全的重要因素,分析了民航维修系统差错的形成和特点,定量评估了民航维修系统差错对民航安全影响的作用形式和影响效果,进一步地探讨了降低民航维修系统差错的方法,在民航维修系统差错未发生前就采取针对性的措施,对于提高安全水平具有重要意义。考虑到民航维修系统差错分析中数据具有模糊和不确定性的特点,以第二代人的可靠性分析方法—CREAM(认知可靠性和失误分析方法Cognitive Reliability and Error Analysis Method)为基础,结合模糊逻辑方法,首次定量研究了民航维修系统差错分析,为采取相关措施降低民航维修系统差错提供了解决方案;进一步以定量化的民航维修系统差错分析为基础,定量分析了民航维修系统差错对维修系统安全的影响。(3)航空机务维修安全指标体系的构建与维修安全评估研究针对民航维修系统的特点,综合分析了影响航空安全的影响因素和每个细节,建立了相应的指标体系;由于采集到的信息具有模糊性和不确定性,研究了基于不确定和模糊信息的民航维修系统的安全评估。综合考虑影响民航维修系统安全的各种因素,包括维修工作、维修管理和维修技术与方法等,首次建立了多层次、多维度的民航维修系统安全评估指标体系,为客观评估机务维修安全水平提供支持。在分析当前主要安全评估方法基础上,采用信息融合方法与基于不确定信息的多属性决策方法,并将综合赋权与变权方法相结合,建立了民航维修系统安全评估模型,充分利用数据,避免了权重赋值对评估结果的影响,该方法科学且具有可操作性。通过对比分析,结果表明,本文提出的方法,较传统的安全评估方法更加准确、可信度高。(4)民航维修系统安全分析和安全监测研究安全分析是发现问题、消除事故隐患的重要手段。本文借鉴人工智能领域的最新研究成果,将Bayes网络方法引入到航空安全分析中,实现了对复杂民航维修系统安全问题的分析。在综合比较当前复杂系统安全分析方法的基础上,建立了基于Bayes网络的民航维修系统安全分析模型,该模型具有动态分析当前安全状态、逆向判断系统危险源的功能,为及时发现安全隐患、采取针对性的预防措施提供了决策支持。结合当前民航维修系统的状态,建立了基于动态Bayes网络的民航维修系统安全监测模型,动态设计与调整监测的内容和环节,实现了民航维修系统安全与效率的有机结合。本文提出的方法在国内某航空公司得到具体应用和实践,取得了良好的应用效果,验证了本文提出方法的有效性。更多还原

【Abstract】 As one of the key industries in the national economic and social development, civil aviation is a high-tech, high-investment, and high-risk industry. Safety is an eternal theme for civil aviation. The ability and level of safety management essentially affects and restraints the development of civil aviation. The research on the civil aviation safety management techniques and methods which are consistent with the characteristics of China and the sustainable safety framework of civil aviation has important significance for increasing the civil aviation safety level and realizing the sustainable civil aviation safety. Aiming at the key and difficult problems of the civil aviation safety analysis and management, the main research contents and innovations of the dissertation are as follows:Firstly, the civil aviation management and the sustainable safety techniques and methods for civil aviation have been studied. The civil aviation system is a huge and complex system. Many factors which have complex logic relationship among themselves affect the safety of civil aviation system. To manage the safety of the civil aviation system is very difficult task. If the theories and methods of system engineering are applied into the civil aviation safety system, and the safety management system which consists of risk identification, safety analysis, safety assessment, safety prediction and safety control could be realized, then the latent dangers in the civil aviation safety system can be found, the benefit and efficiency of the civil aviation safety management system can be improved. In order to realize long term safe operation of civil aviation, a sustainable safety management framework is established, some integrated sustainable safety management techniques and methods of system safety management, process safety management, details safety management, information safety management and aviation safety culture management are studied. Secondly, the errors analysis in engineering maintenance of civil aviation is studied. Engineering maintenance is an important factor for civil aviation safety. The causes and characteristics of errors in the engineering maintenance are analyzed and their impact on civil aviation safety is quantitatively assessed. The approaches to reduce maintenance errors and the appropriate measures to prevent maintenance errors from happening are studied in order to improve safety. Appropriate measures that are adopted before errors happen have significance for increasing the safety level of civil aviation. Considering the fuzzy and uncertain characteristics of errors, the second generation of human reliability analytical method– CREAM (Cognitive Reliability and Error Analysis Method) is applied with fuzzy logic method. Errors in engineering maintenance of civil aviation are quantitatively studied for the first time. It provides the solutions for taking proper measures to reduce maintenance errors. On the basis of the quantitative analysis of errors in maintenance of civil aviation, the quantitative effect of errors in engineering maintenance of civil aviation on the safety of maintenance system is studied.Thirdly, the aviation engineering maintenance safety performance evaluation system is established and the maintenance safety assessment is studied. Aiming at the characteristics of engineering maintenance system of civil aviation, the factors which influence the civil aviation safety are analyzed and the safety performance evaluation system of engineering maintenance system is built. With the fuzzy and uncertain data, the safety assessment of civil aviation maintenance system is studied. Comprehensively considering all factors which have effects on safety of civil aviation, such as maintenance engineering, maintenance management, and maintenance techniques, a multi-level, multi-dimensional safety assessment performance evaluation system for civil aviation engineering maintenance is established for the first time to provide support for impersonally assessing safety level of civil aviation. On the basis of analyzing the present safety assessment methods, data fusion methods and multiple attributes decision methods based on uncertain information are adopted to build the maintenance safety assessment model of civil aviation. Combining comprehensive weighting method and variable weighting method, the influence of weighting on assessment results is alleviated. The safety assessment method is scientific and feasible. The comparison analysis results show that the proposed method is more accurate and more reliable than the traditional safety assessment methods.Fourthly, safety analysis and safety monitoring of civil aviation engineering maintenance are studied. Safety analysis is one important means to find problems and avoid potential accidents. The research results in the field of artificial intelligence, such as Bayes network, are applied to safety analysis of civil aviation. Safety analysis of the complex civil aviation engineering maintenance system is realized. Based on comparisons among the current safety analysis methods of complex system, Bayes network-based safety analysis model of civil aviation engineering maintenance is built. With the help of the model, the present safety status can be analyzed dynamically and the dangerous source can be judged conversely. The model provides support for finding potential safety hazard in time and taking effective preventive measures. Considering the present situation of civil aviation engineering maintenance system, dynamic Bayes network-based safety monitoring model of civil aviation engineering maintenance system is developed. The monitored objects can be designed and adjusted dynamically. Safety and efficiency of the civil aviation engineering maintenance system are improved effectively.The proposed methods in the dissertation have been put into practice in a domestic airline company and have obtained some good effects, which show the effectiveness of the methods.更多还原