基于组合模型的建筑触电事故人因失效分析

为探究触电事故人因关联性,提出一种基于决策试验和评价实验(DEMATEL)-解释结构模型(ISM)-网络层次分析(ANP)的组合模型,对人因失误事故进行失效分析,客观确定各人因影响因素的重要程度。利用认知可靠性和失误分析方法(CREAM)对2012-2021年的192起建筑触电事故进行人因提取并分类,通过关联规则挖掘确定影响因素之间的相关性,生成置信度矩阵,再利用DEMATE-ISM明晰影响因素之间的层级作用路径与网络结构,构建多级递阶结构模型,最后通过ANP对模型进行权重的确定。结果表明,具有较高权重的不完善的安全防护设施管理是根层影响因素,权重最高的决策失误、错过观察为表层直接影响因素,最易形成人因失误事故,必须对其进行严格把控,预防事故的发生。

Human-caused Failure Analysis of Building Electric Shock Accidents Based on Combination Model

To explore the human-related correlation of electric shock accidents, a combined model based on Decision Experiment and Evaluation Experiment (DEMATEL)-Interpretive Structure Model (ISM)-Analytical Hierarchy Network (ANP) is proposed to analyze the failure of human error accidents, objectively, and determine the importance of each individual factor. Using the Cognitive Reliability and Error Analysis Method (CREAM) to extract and classify the human factors of 192 construction electric shock accidents from 2012 to 2021, determine the correlation between the influencing factors through association rule mining, generate a confidence matrix, and then use DEMATE-ISM clarifies the hierarchical action path and network structure between influencing factors, builds a multi-level hierarchical structure model, and finally determines the weight of the model through ANP. The results show that imperfect safety and security facilities management with high weight is the root level influence factor, and the highest weight of decision errors and missed observations are the surface level direct influence factors, which are most likely to form human error accidents and must be strictly controlled to prevent accidents.