基于模糊规则库Fine-Kinney方法的沿海水域碰撞事故人为失误风险分析

Human error risk analysis of coastal water collision accidents based on Fine-Kinney method of fuzzy rule base

引入模糊规则库和传统Fine-Kinney理论相结合的方法对沿海水域碰撞事故人为失误风险进行分析.根据各因子对沿海水域碰撞事故贡献度的分层结果,贡献度值随着管理规定的细化程度降低和船员判断难易程度提高而增加.利用A priority算法对浙江沿海2005—2014年424起海上事故样本进行关联规则挖掘,筛选碰撞事故为前项且人为失误为后项关联规则.利用统计学软件对该方法求出的各因子贡献度值与关联规则中置信度值进行比对分析,证明该方法分析结果基本符合实际情况,可用于海事风险分析研究.该方法首次应用于海事领域,可对单个风险因子进行独立评价.

The human error risk in collision accidents in coastal waters was analyzed by introducing a fuzzy rule base and traditional Fine-Kinney theory. According to the stratified results of contribution degree of each factor to collision accidents in coastal waters, the contribution value increases with the decrease of the refinement of management regulations and the increase of the difficulty of crew judgment. A priority algorithm was used to mine association rules for 424 samples of maritime accidents in Zhejiang coastal area from 2005 to 2014, and the collision accidents were selected as the former and human errors as the latter. Statistical software was used to compare and analyze the contribution value of each factor and confidence value in association rules, and it is proved that the analysis result of the proposed method is basically in line with the actual situation, which can be used in maritime risk analysis. For the first time, this method has been applied in the field of maritime affairs and can independently evaluate a single risk factor.