Association analysis of accident factors in petrochemical storage tank farms

In order to identify and clarify the association between the factors leading to accidents in a petrochemical tank area, this study analyzes investigation reports of 212 petrochemical tank farm accidents and combines this with the “association rule” mining and science related to complex networks. The main risk factors are determined and a risk factor data set is constructed; 75 association rules are extracted from the factor data set based on the Apriori algorithm. Then the obtained association rules are used to construct an accident factors network of the petrochemical storage tank area, and the topology characteristics of the network are further analyzed to reveal the importance of factors. Factors with large node degree, betweenness, and clustering coefficients are obtained, such as “violation of operating regulations”, “high concentration of flammable gas in the air”, “lack of experience and professional skills”, etc. These factors play an important role in the formation and development of accidents. The results also show that the accident cause network of the petrochemical storage tank area has a small average shortest path length and a large cluster coefficient, indicating a relatively close connection between the accident factors. The contributions of this study is not only extracting the hidden relationships among contributory factors to tank farm accidents using association analysis, but also revealing which factors are more important for the tank farm safety through the complex network.     

水电工程高危作业风险影响因素的SEM模型

为探索水电工程高危作业的事故致因机理,耦联分析水电工程施工风险管理过程,从"施工企业影响"、"施工安全管理"、"施工人员行为"、"现场作业相关因素"等方面归纳总结水电工程施工事故风险要素及构成体系;在此基础上,通过问卷调查收集基础数据并定义研究变量属性,抽象风险与影响因素关系的概念模型,建立与检验高危作业风险的SEM模型;然后,以修正指数MI为判据,修正和优化SEM模型。研究结果表明:安全管理制度、教育培训、职业健康环境、个人决策能力对水电高危作业风险具有显著影响,是风险监管的重点,同时也启示风险管理者应深入贯彻"以人为本"的理念,加强"顶层制度设计"。     

综合安全评价(FSA)方法

为提高国际海运安全管理的决策水平,国际海事组织(IMO)接受并倡导由英国海运界提出的综合安全评价(FSA)方法,笔者对FSA方法作了本质性分析和应用性研究。阐述了FSA的由来,5个组成步骤及相互关系和流程,揭示了FSA与风险管理和安全系统工程的渊源;介绍了危险识别的7种工具;对步骤二“风险评估”重点引入了半定量分析方法和定量分析方法,并用集合论观点讨论了事故频率和后果严重度的计算方法;介绍了风险控制方案的“成本与效益评估”的数学模型;归纳了FSA方法的优缺点。指出FSA是一种集风险评价和成本/效益评估于一体,兼顾技术性与经济性,可广泛兼容具体评价方法和普遍适用各类风险评价的框架性方法,但具体应用需要有效整合适用的定性和定量方法。     

基于贝叶斯网络的危化品爆炸事故电力系统风险评估模型

为提高危化品爆炸事故电力应急预警的准确性,建立基于贝叶斯网络的危化品爆炸事故电力系统风险评估模型.基于危化品爆炸事故电力应急典型情景分析,建立综合考虑突发事件、承灾载体和应急管理等风险因素的贝叶斯网络结构.应用概率刻画风险因素信息的不确定性及其相互影响,定量分析事件后果.结合一般条件和典型情景等的应用实例,分析评价方法...     

A study of maintenance-related major accident cases in the 21st century

This paper is based on a review of 183 detailed, major accident investigation and analysis reports related to the handling, processing and storage of hydrocarbons and hazardous chemicals over a decade from 2000 to 2011. The reports cover technical, human and organizational factors. In this paper, the Work and Accident Process (WAP) classification scheme is applied to the accident reports with the intention of investigating to what extent maintenance has been a cause of major accidents and what maintenance-related causes have been the most frequent.The main objectives are: (1) to present more current overall statistics of maintenance-related major accidents, (2) to investigate the trend of maintenance-related major accidents over time, and (3) to investigate which maintenance-related major accident causes are the most frequent, requiring the most attention in the drive for improvement.The paper presents statistical analysis and interpretation of maintenance-related major accidents’ moving averages as well as data related to the types of facility, hazardous substances, major accidents and causes. This is based on a thorough review of accident investigation reports.It is found that out of 183 major accidents in the US and Europe, maintenance was linked to 80 (44%) and that the accident trend is decreasing. The results also show that “lack of barrier maintenance” (50%), “deficient design, organization and resource management” (85%) and “deficient planning/scheduling/fault diagnosis” (69%) are the most frequent causes in terms of the active accident process, the latent accident process and the work process respectively.     

基于关联规则的跑道着陆事件演化网络构建研究*

为了提高航空运输系统安全性,针对航空器着陆过程中冲偏出和重着陆事件的关联机理开展研究。以84起跑道着陆事故调查报告为样本,利用Python分词工具提取风险因素并建立跑道着陆事件数据集,应用Apriori算法挖掘风险因素间关联规则,基于复杂网络理论构建该类事件演化网络,分析社团结构和拓扑特征。研究结果表明:本文所构建的网络密度为0.143,网络较为紧密,风险因素间强关联;网络社团结构显著,符合飞行操纵、运行状态、运行环境、不利条件的分类特点;网络关键节点“航线训练”、“接地速度大”综合值分别为0.599 8、0.589 0。研究结果可为飞行安全研究提供新思路。     

Comparison of techniques for accident scenario analysis in hazardous systems

In this paper, three accident scenario analysis techniques are presented and compared regarding their efficiency vs. the demanded resources. The complexity of modern industrial systems has prompted the development of accident analysis techniques that should thoroughly investigate accidents. The idea of criteria classification to fulfill this requirement has been proposed by other researchers and is examined here too. The comparison is done through the application of Event Tree analysis, Fault Tree analysis and Petri Nets technique—two relatively simple and a more demanding methodology—on the same hazardous chemical facility in view of analyzing an accident scenario of a hazardous transfer procedure. Accident scenario analysis techniques are essential not only in learning lessons from unfortunate events in the chemical industry but also in preventing the occurrence of such events in the future and in communicating risk more efficiently.     

论本质安全

考查“本质安全”一词的出处 ;介绍了几种“本质安全”的论述 ;讨论了“以人为本”和“人本管理”与本质安全 ;从政治和哲学的高度 ,论述了政治、现象与本质的内涵与关系 ;探讨了美国及发达国家是否有“本质安全”;分析了现代资本家重视职业安全卫生的实质。笔者对以上问题阐明了自已的观点 ,并对一些提法给以评述。     

A methodology to model causal relationships on offshore safety assessment focusing on human and organizational factors

INTRODUCTION: Focusing on people and organizations, this paper aims to contribute to offshore safety assessment by proposing a methodology to model causal relationships. METHOD: The methodology is proposed in a general sense that it will be capable of accommodating modeling of multiple risk factors considered in offshore operations and will have the ability to deal with different types of data that may come from different resources. Reason's "Swiss cheese" model is used to form a generic offshore safety assessment framework, and Bayesian Network (BN) is tailored to fit into the framework to construct a causal relationship model. The proposed framework uses a five-level-structure model to address latent failures within the causal sequence of events. The five levels include Root causes level, Trigger events level, Incidents level, Accidents level, and Consequences level. To analyze and model a specified offshore installation safety, a BN model was established following the guideline of the proposed five-level framework. A range of events was specified, and the related prior and conditional probabilities regarding the BN model were assigned based on the inherent characteristics of each event. RESULTS: This paper shows that Reason's "Swiss cheese" model and BN can be jointly used in offshore safety assessment. On the one hand, the five-level conceptual model is enhanced by BNs that are capable of providing graphical demonstration of inter-relationships as well as calculating numerical values of occurrence likelihood for each failure event. Bayesian inference mechanism also makes it possible to monitor how a safety situation changes when information flow travel forwards and backwards within the networks. On the other hand, BN modeling relies heavily on experts' personal experiences and is therefore highly domain specific. IMPACT ON INDUSTRY: "Swiss cheese" model is such a theoretic framework that it is based on solid behavioral theory and therefore can be used to provide industry with a roadmap for BN modeling and implications. A case study of the collision risk between a Floating Production, Storage and Offloading (FPSO) unit and authorized vessels caused by human and organizational factors (HOFs) during operations is used to illustrate an industrial application of the proposed methodology.     

基于数据挖掘的化工生产事故致因主题抽取

为充分挖掘化工生产事故数据中的有效信息和潜在规律,提高对化工事故认知水平,针对某化工集团2010—2016年共1 578起事故数据,利用社会网络分析等方法揭示事故要素间的关联关系;运用潜在狄利克雷分配（LDA）模型进行事故聚类,并抽取到5个事故致因主题。研究结果表明:LDA主题模型等数据挖掘技术能有效挖掘大量事故数据中的潜在信息;5个事故致因主题中,4个涉及到人因或组织层面的缺陷;员工注意力不集中和现场风险管理不足这2个致因主题间具有较强相关性;员工注意力不集中、现场风险管理不足以及设备问题是导致事故发生的主要原因。     

省级生产安全应急救援体系建设方案探讨

省级生产安全应急救援体系是国家生产安全应急救援体系的重要组成部分。为了使省级生产安全应急救援体系的建设方案具有实用性、先进性、经济性和可操作性 ,笔者通过调研许多地方的经济现状、重大危险源和现有的生产安全应急救援资源 ,提出了省级应急救援体系的基本框架和体系在应急救援情况下的响应程序。要在省、市 (地区 )、县三级建立起权威的应急救援指挥机构 ,才能使应急救援体系实现高度的统一指挥 ,灵敏高效 ;充分地整合和利用现有各种救援力量和通讯系统 ,实现资源共享 ,避免重复建设 ;促进“110” ,“119” ,“12 0” ,“12 2”等通讯网络联动 ,并逐步应用GPS ,GIS等资源 ,建立先进的动态跟踪监测 ,可视的信息系统 ;应急救援队伍由3部分组成 ,即政府主办的公共应急救援队伍 (现有 )、政府资助由行业或集团公司主办的专业应急救援队伍、企业自备的应急救援队伍 ;同时对体系中各种机构的组成及其主要职责提出了设想。     

电梯PAR安全信息流管理模型构建及应用

为从安全信息视角分析电梯全生命周期过程中事故致因因素,并在实践中提升电梯安全管理水平,基于PAR事故致因模型开展电梯安全信息管理研究。以安全信息为主线,安全感知-安全分析-安全应答(PAR)安全行为链为关键点,建立电梯PAR安全信息流管理模型。基于电梯PAR安全信息流管理模型的逻辑架构建立电梯信息综合监管服务体系,将模型的信息服务形式具体化。并运用电梯PAR安全信息流管理模型对典型电梯事故进行分析。研究结果表明：电梯PAR安全信息流管理模型能够全面分析各层级失误模式,提升各主体协同治理能力。电梯信息综合监管服务体系可打破电梯领域信息“孤岛”现状,对智慧电梯建设具有重要理论指导意义。     

事故调查报告中事故原因分析的改进*——基于企业事故案例学习视角

为进一步提高我国事故调查报告的质量,更好地满足企业对事故案例学习的需要,通过案例分析事故调查报告中事故原因部分存在的问题及其原因,并在此基础上提出采用事故致因理论来改进事故调查报告的思路;应用24Model改进天嘉宜化工有限公司“3·21”特别重大爆炸事故的事故原因分析,以此来验证本文思路的可行性和有效性。研究结果表明,使用事故致因理论能够更加科学地归纳事故原因,增强措施的针对性和可操作性,可有效改善事故调查报告的质量。     

基于ANP的建筑安全管理绩效评价框架研究

基于政府A部门提供的502起建设工程事故调查报告,综合应用比较分析、数据挖掘、网络层次分析法(ANP)和超级决策(SD)决策软件,建立建筑安全管理绩效评价框架,包括6个一级指标和24个二级指标;通过分析各因素之间的关系,修正各个因素的权重,得出事故致因的传递路径。根据和其他偏定性研究的比较还可以发现,基于数据挖掘的ANP和SD的综合运用,更利于对真实情况进行研究模拟,在应对较大建设工程事故风险时更具有针对性和前瞻性。     

高危行业事故应急科技服务实用性平台研究

针对当前我国高危行业应急平台建设中存在事故应急救援决策支撑弱、应急救援服务能力差的缺陷,以“江苏省高危行业事故应急科技服务平台”项目为背景,提出高危行业事故应急科技服务平台的总体建设架构;并以此为基础提出建立应急信息共享,预案可视化与推演,事故后果分析模拟,应急救援决策支持,应急咨询、培训、评估,事故现场快速监测六大子平台。以期利用VR、GIS、人工智能等先进技术,从功能、系统、体系三个层面完善高危行业事故应急科技服务平台建设内容,从而为政府门提供应急决策支持、为企业和社会公众提供应急科技服务。     

谈职业健康安全管理体系审核的有效性

笔者从审核人日、抽样的合理性、审核员能力、审核过程的有效性等4个方面讨论了职业健康安全管理体系(OHSMS)审核的有效性。从要素的角度,介绍国家标准GB/T28001—2001中“对危险源辨识,风险评价和风险控制的策划”,“法律法规及其他要求”,“职业健康安全管理方案”,“结构和职责”,“运行控制”,“绩效测量与监测”,“事故事件,不符合,纠正与预防措施”等7个要素审核过程的有效性。可为认可认证机构判断OHSMS审核的有效性提供一定的参考。     

基于Apriori算法的易燃易爆危险化学品储运火灾爆炸事故关键致险因素的挖掘

导致易燃易爆危险化学品储运火灾爆炸事故的因素繁多、关系复杂,挖掘关键致险因素是减少管理成本、提高防控效率的关键。研究了200例事故等级为较大事故以上的易燃易爆危险化学品储运火灾爆炸事故的原因,采用事故树分析法建立易燃易爆危险化学品储运火灾爆炸事故树,并运用频度统计法遴选出致险因素;在此基础上,建立基于Apriori算法的关联规则模型进行数据挖掘,共得到14个关键致险因素。通过对关键致险因素与易燃易爆危险化学品储运火灾爆炸事故之间关联规则的分析表明,关键致险因素与事故之间存在强关联规则,单一关键致险因素或其组合的存在必将导致事故的发生,为实现危险化学品储运精准化安全管理提供参考。     

Prerequisites for learning from accident investigations – A cross-country comparison of national accident investigation boards

In this paper railway accident investigation reports issued by the national accident investigation boards in three Scandinavian countries during a 2-year period have been systematically studied. Content analysis of attributed causes in these reports reveals a considerable emphasis on physical processes, actor activities and equipment (the microlevel). Much less attention is paid to organisational factors (the mesolevel) and conditions related to regulators, associations and government (the macrolevel). This means that lessons will primarily be learned about aspects at the lower of these levels. Interviews show that the factors emphasised in investigation reports typically reflect the competences and experiences of the investigators, i.e. they are inclined to focus on areas of their own expertise. Since failures at the microlevel in many cases merely are symptoms of trouble at higher levels, it is argued that competence among investigators that supplements entirely technical or operational backgrounds are necessary for enabling deeper understanding of the factors leading to accidents. One possible way for achieving this is the creation of multi-modal investigation boards that provide a number of potential advantages, such as increased access to specialist competences that are shared between different sectors. Although a multi-modal approach to some degree has been adopted in all three countries, interviews reveal that these positive effects do not emerge automatically. It can therefore be concluded that multi-modal investigation boards offer a number of possible advantages, but only when these synergies are fully exploited can they provide a potential for more effective learning from accidents.     

事故调查分析方法与技术述评

根据事故的原因结果模型、过程模型、能量模型、逻辑树模型、SHE管理模型将事故调查分析技术方法归纳为5大类,介绍了广泛使用的事故调查分析技术方法的特点,并按一定的标准进行分析和比较。分析表明:每种方法都有其不同的应用领域和优缺点,应针对不同领域问题选用合适的分析方法,多种调查技术组合分析适用于复杂事故的调查;考虑事件次序及其影响因素的调查方法,便于提出预防事故再次发生和减少风险的建议,图表阐述有利于鉴别信息缺陷,并使调查组之间的沟通有力;笔者在调查过程中注重于非正常分析方法得出的结果,宜采用更为先进的事故分析技术方法以保证事故调查的客观性。