College of Mechanical and Transportation Engineering, China University of Petroleum (Beijing Campus), Beijing 102249, PR China.
College of Mechanical and Transportation Engineering, China University of Petroleum (Beijing Campus), Beijing 102249, PR China.
J Hazard Mater. 2014 Aug 15;278:529-38. doi: 10.1016/j.jhazmat.2014.06.034. Epub 2014 Jun 25.
Fire and explosion accidents of steel oil storage tanks (FEASOST) occur occasionally during the petroleum and chemical industry production and storage processes and often have devastating impact on lives, the environment and property. To contribute towards the development of a quantitative approach for assessing the occurrence probability of FEASOST, a fault tree of FEASOST is constructed that identifies various potential causes. Traditional fault tree analysis (FTA) can achieve quantitative evaluation if the failure data of all of the basic events (BEs) are available, which is almost impossible due to the lack of detailed data, as well as other uncertainties. This paper makes an attempt to perform FTA of FEASOST by a hybrid application between an expert elicitation based improved analysis hierarchy process (AHP) and fuzzy set theory, and the occurrence possibility of FEASOST is estimated for an oil depot in China. A comparison between statistical data and calculated data using fuzzy fault tree analysis (FFTA) based on traditional and improved AHP is also made. Sensitivity and importance analysis has been performed to identify the most crucial BEs leading to FEASOST that will provide insights into how managers should focus effective mitigation.
钢质油库火灾爆炸事故（FEASOST）在石油和化工生产及储存过程中时有发生，对生命、环境和财产造成了巨大的破坏。为了开发一种定量评估 FEASOST 发生概率的方法，本文构建了一个 FEASOST 的故障树，确定了各种潜在的原因。传统的故障树分析（FTA）如果所有基本事件（BEs）的失效数据都可用，则可以进行定量评估，但由于缺乏详细数据以及其他不确定性，这几乎是不可能的。本文尝试通过基于专家启发的改进层次分析法（AHP）和模糊集理论的混合应用来进行 FEASOST 的 FTA，并对中国某油库的 FEASOST 发生概率进行了估计。还比较了基于传统和改进 AHP 的模糊故障树分析（FFTA）的统计数据和计算数据。进行了敏感性和重要性分析，以确定导致 FEASOST 的最关键的 BEs，这将为管理人员提供如何集中有效缓解的见解。
