摘要
简要介绍了传统的人因飞行事故致因理论及其模型,特别是HFACS航空人因事故致因分析体系。借鉴美国海航和国内外民航的人因分析分类体系,结合空军实际,增加了"组织计划"致因要素,调整了"组织管理"致因要素,提出了一种改进的HFACS航空人因事故致因分析模型。利用系统时效、质量和结构熵对两种模型的有序度进行了评价,选取5种不同偏好的权重值计算了模型的有序度。改进模型的结构熵小,而质量较大,因此是相对较优的模型。运用因子分析(FA),使用SPSS 18.0统计分析软件包对模型结构体系中的16个指标进行了数据分析。对各指标数值运用Z-SCORE法进行了标准化处理,对公因子进行了方差最大化旋转,以累计方差贡献率大于85%为标准,选取了5个公因子代替原有16个指标对人因事故进行分析结果表明,提取的评价因子均为单维指标,不存在指标冗余现象,从而验证了改进模型的合理性和可行性。
According to the aviation accident statistical results, more and more aviation accidents have been caused by human manipulation errors. Therefore, it is necessary to pay enough attention to the study of aviation accidents caused by human manipulation factors so as to be able to evaluate and analyze adequately the human-oriented factors and in turn take forceful measures in advance so as to ensure the air- flight safety. For this purpose, this paper would like to put our re- search focus on the following aspects： First of all, we would like to introduce some traditional accident causation theories and models about human factors in flight accident study, especially, Wang Yong gang＇ model based on the HFACS aviation accident causation analy- sis. Secondly, based on the American Naval Aviation and our domes- tic civil aviation human factors analysis classification system, so as to build up an improved HFACS aviation accident causation model, to which it is necessary to add up the organizational factors, the adjust- ment and regulation factors, in addition to the administrative manage- ment factors. Thirdly, the analysis of the aging problem of the ma- nipulating systems, their quality and structure entropy formula, in- cluding the evaluation of the degree of system order, the selection of the aging entropy, the maximum timehness entropy, the quality en-tropy and the maximum quality entropy, the calculated structure of the degree of order to verify structure in the right to re-bias stability under the conditions. Through comparison, the structure entropy of the improved model should be lower, whereas the aging and model quality should be more stressed. And, so is it the better structural model. And, finally, the improved model should be analyzed and confirmed by the method of factor analysis （FA）. In addition, we have adopted SPSS 18.0 statistical analysis software to analyze the data of 16 index model structure system, and standardized the index values by using Z - SCORE method. The common factors is rotated to variance maximization. Since the variance contribution rate can be greater than 85 %, we have chosen five common factors to replace the original 16 indicators for the accident evaluation. From what is men- tioned with our improved model, the evaluation factors can all be treated all through the single dimension indexes, so that there will be no index redundancy, which can prove the feasibility of our improved model.
出处
《安全与环境学报》
CAS
CSCD
北大核心
2013年第5期229-233,共5页
Journal of Safety and Environment
基金
国家自然科学基金项目(71171199)
关键词
安全工程
航空事故
人的因素
事故致因模型
结构熵
因子分析
safety engineering
aviation accident causation model
analysisaccident
human factors
structure entropy
factor
