目前,我国对事故事件资源的利用水平仍处于较低阶段,难以充分挖掘事故数据中的潜在规律。为深入研究油气行业事故致因因素,提升事故预防能力,本文基于HFACS (人为因素分析与分类系统)模型和文本挖掘技术,对国内某油田公司2016~2022年间...目前,我国对事故事件资源的利用水平仍处于较低阶段,难以充分挖掘事故数据中的潜在规律。为深入研究油气行业事故致因因素,提升事故预防能力,本文基于HFACS (人为因素分析与分类系统)模型和文本挖掘技术,对国内某油田公司2016~2022年间发生的204起生产建设事故报告进行分析,构建适用于油气行业的HFACS-OGI事故致因分类模型,并采用Apriori关联规则算法挖掘事故因素之间的强关联模式。研究结果表明:在HFACS-OGI模型中,“组织影响”是最关键的致因层级;通过关联规则分析揭示了油气生产事故致因之间的多层级确定性关联和系统性漏洞;基于研究结果,本文提出针对性的事故预防策略,以期为油气生产企业提供相关参考。At present, the utilization level of accident event resources in China is still at a relatively low stage, making it difficult to fully explore the potential patterns in accident data. In order to conduct in-depth research on the causes of accidents in the oil and gas industry and enhance accident prevention capabilities, this paper analyzes 204 production and construction accident reports that occurred in a domestic oilfield company from 2016 to 2022 based on the HFACS (Human Factors Analysis and Classification System) model and text mining technology. A HFACS-OGI accident cause classification model suitable for the oil and gas industry is constructed and the Apriori association rule algorithm is used to mine strong correlation patterns between accident factors. The research results indicate that in the HFACS-OGI model, “organizational influence” is the most critical causal level;through association rule analysis, multi-level deterministic associations and systematic loopholes between the causes of oil and gas production accidents have been revealed;based on the research results, this article proposes targeted accident prevention strategies in order to provide relevant references for oil and gas production enterprises.展开更多
文摘目前,我国对事故事件资源的利用水平仍处于较低阶段,难以充分挖掘事故数据中的潜在规律。为深入研究油气行业事故致因因素,提升事故预防能力,本文基于HFACS (人为因素分析与分类系统)模型和文本挖掘技术,对国内某油田公司2016~2022年间发生的204起生产建设事故报告进行分析,构建适用于油气行业的HFACS-OGI事故致因分类模型,并采用Apriori关联规则算法挖掘事故因素之间的强关联模式。研究结果表明:在HFACS-OGI模型中,“组织影响”是最关键的致因层级;通过关联规则分析揭示了油气生产事故致因之间的多层级确定性关联和系统性漏洞;基于研究结果,本文提出针对性的事故预防策略,以期为油气生产企业提供相关参考。At present, the utilization level of accident event resources in China is still at a relatively low stage, making it difficult to fully explore the potential patterns in accident data. In order to conduct in-depth research on the causes of accidents in the oil and gas industry and enhance accident prevention capabilities, this paper analyzes 204 production and construction accident reports that occurred in a domestic oilfield company from 2016 to 2022 based on the HFACS (Human Factors Analysis and Classification System) model and text mining technology. A HFACS-OGI accident cause classification model suitable for the oil and gas industry is constructed and the Apriori association rule algorithm is used to mine strong correlation patterns between accident factors. The research results indicate that in the HFACS-OGI model, “organizational influence” is the most critical causal level;through association rule analysis, multi-level deterministic associations and systematic loopholes between the causes of oil and gas production accidents have been revealed;based on the research results, this article proposes targeted accident prevention strategies in order to provide relevant references for oil and gas production enterprises.
