地铁施工事故统计分析与研究*

为了解我国城市地铁施工事故规律特征,降低新建地铁线路施工事故发生频率,统计分析2011—2020年全国31个省市自治区（不包括港澳台）地铁施工事故,归纳事故发生时间、事故类型、施工工法的分布特征,基于灰色关联分析法（GRA）对地铁施工事故致因进行分析。研究结果表明:地铁施工事故季节性特征突出,3,8,11月份呈多发态势;科学技术进步和规章制度完善是保证我国轨道交通运营里程高速增长和施工事故率降低的有效手段;事故类型以坍塌为主,且事故造成影响最大;违章作业、设备设施缺陷、地质环境不良和安全监管未贯彻落实分别为“人、物、环、管”4个维度的关键致因因子。     

Port safety and the container revolution: A statistical study on human factor and occupational accidents over the long period

Several factors can affect occupational accident frequency, namely economical factors, technologies used (low automation, discontinuous operating) job design, organization of work/environmental conditions and human factors. In particular, technological advances in industrial activities can give rise to improvement in productivity and in occupational health and safety, but not necessarily simultaneously. The beginning of the container transport dates back to 50 years ago, but while containerization changed everything, from ships and ports to patterns of global trade, its impact on work injuries was not explored at all. The aim of this paper is to investigate the relationship between work organization, job experience, productivity and occupational accidents, from the starting of the container expansion to nowadays, considering Genoa port (Italy), one of the largest of the Mediterranean Sea. In order to minimize possible reporting biases, such as underreporting or reclassification to a lower level of severity, injury statistics are elaborated starting from data collected directly on-site, from internal accident or medical-aid reports. An in-depth statistical analysis on occupational injuries in the years 1980–2006 is carried out, with reference to frequency indexes, mechanism of injury and material causes. The increase of container-ships traffic and, consequently, the sharp change in port infrastructure involved a rapid modification also in the work organization, with particular reference to the number and characteristics of workforce (decrease from 5783 to nearly 1000 employees and increase of low experience workers from 28% to 74%). The striking high percentage increase of young or low experienced workers in handling container (and performing correlated new tasks) caused a remarkable increase of the risk for occupational injuries. In the studied port, we recorded an increase of the frequency index (injuries per hundred thousand hours worked) from 13.0 to 29.7. It results that the increased expansion of shipping container utilization is not connected to a correspondent human factor safety implementation. Main risk factors are pointed out, revealing an increase of accidents due to transport vehicle (+8.3%) and a reduction of accidents caused by substance or materials (?4.5%). These factors show a statistical significant correlation with the new job tasks. Consideration of these findings may enable managerial solutions and workplace organization interventions for the prevention of injuries and safety performance improvement in port activities.     

基于Tripod-Beta模型的船舶火灾事故风险分析

为深入分析船舶火灾事故风险因素及其后果产生的影响,通过分析1991-2017年全球船舶火灾事故调查报告,从人员、管理、船舶设备、货物、环境5个方面对船舶火灾影响因素进行识别研究;采用三脚架事故致因模型(Tripod-Beta model),构建考虑安全栅的船舶火灾事故情景演化模型,识别船舶火灾关键影响因素;并在样本量较少的情况下,采用信息扩散理论计算船舶火灾发生率;最后,利用布尔函数和风险矩阵,对船舶火灾事故风险进行评价研究。结果表明:船员不安全行为和船舶设备表面过热、设备短路是船舶火灾事故的关键风险因素;事故后果链中安全栅遭到破坏时,船舶火灾风险处于不希望发生范围内。该方法能有效评估船舶火灾风险的等级,满足海事管理部门的监管工作需求。     

Carbon dioxide pipelines: A statistical analysis of historical accidents

Carbon dioxide pipeline is an essential carrier in carbon capture, utilization, and storage (CCUS). Statistically revealing the accident rate and risk of carbon dioxide pipelines is conducive to integrity management. Based on 112 accident records collected from Pipeline & Hazardous Materials Safety Administration, this work analyzes the frequency, rate, and risk of accidents. In addition, the impact of relevant factors on risk is further discussed. Some primary conclusions are as follows: (1) For carbon dioxide pipelines, the leak is the leading form of accident. Most carbon emissions are generated in the form of leakage, but economic losses are mainly generated in other forms. (2) The pipelines that have been in service for 0–10 years have the highest frequency of accidents and the highest proportion of carbon emissions, but the pipelines that have been in service for 11–20 years have caused the most economic losses. (3) Among the accident causes, the number of accidents caused by equipment failure is the highest, while the economic loss caused by natural force damage is the highest, and the carbon emission caused by material failure is the highest.     

基于动态风险平衡的海洋平台事故连锁风险研究

针对海洋平台事故风险特点,提出动态风险平衡概念,以此建立事故动力模型,并将该模型运用到墨西哥湾"深水地平线"井喷事故。动态风险平衡表征事故动力与事故阻力之间的动态平衡状态,具有动态性和暂时稳定性。事故动力模型以海洋平台可能发生的重大事故为研究对象,从工艺、技术和管理等角度分析事故可能致因和事故发展可能影响因素。该模型首先分析对象的初始事故动力,建立事故连锁风险图,然后计算初始动力发生情况下,传递动力和传递阻力的概率分布,最后提出相应风险控制措施。实例分析表明,基于动态风险平衡建立的事故动力模型能有效分析海洋平台事故连锁风险。     

Bifocal/varifocal spectacles, lighting and missed-step accidents

Good lighting and the correct use of spectacles generally improve vision and presumably reduce the risk of accidents. However, bifocal and varifocal spectacles can increase the risk of misjudging distances when negotiating underfoot hazards. In some circumstances, the portion of the lens used for close work may inadvertently be used where accurate judgement of distance is necessary. This paper reports results from two studies of patient interviews using the Merseyside Accident Information Model; the first a study of accidents which occurred during paid employment (1504 cases) and the second of 1326 accidents mainly in domestic and leisure activities. Underfoot accidents were identified by the first unforeseen event perceived by the patient. The relationship between underfoot events and the type of spectacles worn by the patient at the time of accident was analysed. In both studies there was a significant association between accidents where the first event was ‘missed edge of’ (step) and wearing bifocal/varifocal spectacles. Corporal movements reported in both studies indicated that ‘stepping down’ when wearing bifocal/varifocal spectacles, increased the risk of ‘missed edge’ accidents. In the work environment corporal movements associated with manual handling of loads also increased the risk of ‘missed edge’ accidents when wearing bifocal/varifocal spectacles.     

Road accidents,an occupational risk

Road accidents sustained at work represent between 20% and 40% of work fatalities in most industrialised countries, yet few data on occupational road accident risk factors have been published. A case control study was performed to assess the role of work-related risk factors in the occurrence of occupational road accidents. A preliminary qualitative study was carried out to identify possible occupational factors in occupational road accidents, and to draw up the case control study. Cases were recruited from the Rhône road trauma registry (France), controls from voting lists. A telephone interview was performed. Exposure to road risk was measured as a percentage of work time. One hundred and forty-six cases and 440 matched controls were interviewed. Accident risk was found to increase with exposure. Driving was associated with more difficult working conditions than found in jobs not involving driving. These difficulties, however, were not systematically associated with increased occupational road accident risk. Among factors which still emerge after adjustment for road risk exposure, there are scheduling issues (inflexible schedule organisation, lack of consecutive rest-days, lack of flexibility in performing the work), difficulties of communication with superiors, low seniority in the activity, low educational level and physical constraints at work. This study highlights some possible occupational road accident risk factors. Given the chosen case/control methodology, the findings may be considered as advancing our knowledge of the subject, but need confirmation by further studies.     

Differences in the lethality of occupational accidents in OECD countries

International comparisons of the frequency of occupational accidents are rare because figures reported by each country are influenced by factors other than differences in the occurrence of actual accidents. In this study, we performed an international comparison of occupational accidents which distinguishes between fatal cases and all reported accidents, in order to control for the effects of other factors in accident reporting. Annual data on the numbers of fatal and non-fatal occupational accidents in member countries of the Organisation for Economic Co-operation and Development (OECD) between 1993 and 1998 were obtained from the Labour Statistics Database, which is operated by the International Labour Organization (ILO). The average incidence of all accidents across OECD countries ranged from 17 to 4608 per 100,000 persons in the labor force, whereas the incidence of fatal accidents across OECD countries ranged from 1 to 12 per 100,000 persons in the labor force. Lethality rates per 10,000 total accident cases ranged from 9 (Germany) to 3252 (Turkey). When comparing countries using the same statistical definition of accidents, three or more absence days, the difference in lethality of reported accidents within the same industrial sector is reduced to a sixfold difference (49 for the United Kingdom versus 333 for Ireland). Furthermore, the difference in lethality decreased to about twofold (18 for Spain versus 35 for Czech Republic) in the group using definitions with a criteria of one absent day. The results suggest that reported differences in lethality can reflect artificial administrative factors rather than the harmful factors that cause actual occupational accidents. To make more valid comparison of occupational accident between countries, harmonization of accident statistics is required.     

聚乙烯醇罐区火灾爆炸事故树分析

通过现场的调研与事故树分析相结合的手段对某厂聚乙烯醇车间聚合罐区火灾爆炸事故的危险因素进行了识别与分析.以该罐区可能发生的火灾爆炸事故作为顶上事件,对可能引发顶上事件的21个基本事件及一个条件事件构建事故树,利用最小割集、最小径集及结构重要性计算手段进行事故风险程度分析,从而确定醋酸乙烯暴聚是聚合罐区的首要危险源,而促发醋酸乙烯暴聚的物料长时间停留、气相氧含量过高、温度控制失效、阻聚剂含量不足等四个基本事件是导致聚合罐区火灾爆炸事故的最危险因素.本文对以上聚合罐区发生火灾爆炸事故的风险因素进行详细定性分析,并在此基础上有针对性的提出了相应的安全预防控制措施.同时,该聚合罐区的事故树分析结论也可以为同类别化工单位罐区的日常运行、设计改造、维护保养等工作提供理论依据.     

重大基坑坍塌事故虚拟仿真系统研究

为预防基坑坍塌事故发生、进行重大事故警示教育,以杭州地铁湘湖站北2基坑坍塌事故为素材,研究构建重大基坑坍塌事故虚拟仿真培训系统。对案例事故涉及的设计、施工、环境三大因素的技术问题进行总结,研究了事故链各致险因素及对推动事故发生所起作用,完成仿真系统平台设计和开发工作。系统应用多种制作技术创建工程场景模型,展示深基坑工程复杂的施工工序和空间关系,再现基坑坍塌事故发生发展过程和贴近实际的灾难场景。研究结果表明:仿真系统成果可丰富地下工程教学、施工现场安全教育和安全监察培训的内容和形式。     

Equipment failures and their contribution to industrial incidents and accidents in the manufacturing industry

Accidental events in manufacturing industries can be caused by many factors, including work methods, lack of training, equipment design, maintenance and reliability. This study is aimed at determining the contribution of failures of commonly used industrial equipment, such as machines, tools and material handling equipment, to the chain of causality of industrial accidents and incidents. Based on a case study which aimed at the analysis of an existing pulp and paper company's accident database, this paper examines the number, type and gravity of the failures involved in these events and their causes. Results from this study show that equipment failures had a major effect on the number and severity of accidents accounted for in the database: 272 out of 773 accidental events were related to equipment failure, where 13 of them had direct human consequences. Failures that contributed directly or indirectly to these events are analyzed.     

Factors in offshore planning that affect the risk for major accidents

The purpose of this paper is to systematically analyse a typical planning process in the offshore industry from the perspective of causes of major accidents, with the ultimate aim of identifying factors that affect the risk for major accidents occurring. We first study and describe a typical planning process for offshore oil and gas operations in Norway. Then we analyse a number of theories of major accidents, to see how the different theories and their explanations of causes and contributing factors can be of relevance for future plans and planning processes. Finally, we review accident investigations to search for evidence of how weaknesses in planning processes can contribute to major accidents through the above identified factors. Also, we try to identify any additional factors that have not been recognised through the theoretical review. This provides empirical support for the theoretical basis. Thirteen factors which directly or indirectly can influence the planning process causing a major accident potential are identified. These are exemplified through a review of investigation reports. The paper suggests that planning process should focus more on increasing quality in the plans at an early phase, with examples from incidents, and illustrate the relation between planning quality and potential for major accidents.     

超大城市昼夜交通事故严重程度致因对比分析——以深圳市为例

为研究夜间交通事故严重程度致因,基于深圳市3年3 244起交通事故数据,获取昼夜交通事故分布的时空特征;进一步选取交通事故集聚的南山区、福田区、罗湖区的1 798起交通事故,以交通事故严重程度为因变量,以事故原因、日期、事故形态等10个因素为候选自变量,构建广义有序Logit回归模型,对比分析昼夜不同严重程度交通事故的影响因素。结果表明:路口路段类型、疲劳驾驶、事故日期在夜间模型参数估计值分别为0.493,-0.363,-0.309,而在日间模型表现为不显著,道路路面材料在日间模型参数估计值为-0.232,而在夜间表现为不显著;事故原因、道路横断面渠化方式等因素在日间和夜间所引起交通事故的严重等级均存在较大差异。     

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。     

The use of current risk analysis tools evaluated towards preventing external domino accidents

Risk analysis is an essential tool for company safety policy. Risk analysis consists of identifying and evaluating all possible risks. The efficiency of risk analysis tools depends on the rigueur of identifying and evaluating all possible risks. The diversity in risk analysis procedures is such that there are many appropriate techniques for any circumstance and the choice has become more a matter of taste. In this paper, we examine the risk analysis tools used by 24 chemical plants in Belgium, mainly located in the port of Antwerp, the second largest chemical cluster in the world after Houston, TX, USA. The aim of this paper is to identify the current practice in the chemical industry subject to European Seveso legislation and to examine how the present methods can be integrated to improve safety policy, especially towards preventing major accidents. Moreover, conclusions on the prevention of catastrophic external domino accidents involving several companies are formulated. This paper also gives impulse to investigating cross-company management implementations concerning external domino accident prevention.     

考虑结构特性的化工事故关键致因网络分析*

为揭示化工事故发生机理,考虑事故链中因素的因果关系与事故网络的结构特性,基于复杂网络研究化工事故的关键致因因素,以事故案例中因素与结果为网络节点,因素间的因果关系为边,建立化工事故网络模型,分析网络中节点的度值、结构洞,构建化工事故凝聚网络,分析凝聚节点之间的因果关系。结果表明:在化工事故网络中存在8种致因因素,人的不安全行为与环境的不安全状态占比高达87.5%。通过结构洞分析可有效找出网络中具有较强影响力的潜在致因因素;通过凝聚网络分析,人的不安全行为致使设备的不安全状态与环境的不安全状态连接比例高达29%,存在显著的因果关系。     

Risk of repeat accidents by economic activity in Italy

RationaleThe phenomenon of repeat work-related accidents has not been sufficiently studied.ObjectivesTo evaluate the role of work-related factors in the occurrence of repeat accidents in Italy, using economic activity and the size of the enterprise as proxies and basing the analysis on available administrative data.MethodsThe data, provided by the Italian Workers’ Compensation Authority (INAIL), refer to accidents in the industry and artisan sectors and in the service industry between 1994 and 1999. We selected the data on accidents among 633,735 persons 25–55 years of age with a first accident in 1996–1997, and we determined the occurrence of further accidents within two years in the same economic activity, taking into account job mobility and factors associated with underreporting (geographic area, age, and size of the enterprise). The gender-specific relative risk was computed for each economic activity, given by the ratio between the risk of repeat accidents in a given economic activity and the risk in all of the other activities.ResultsIndividuals with a single accident differed from those with a repeat accident in terms of the distribution by gender, age-class, and severity of injury. The economic activities with the highest relative risk were generally those known to represent a risk of accidents in general: metal ore and coal mining, shipbuilding, the manufacturing of railroad equipment, building construction, road and railway construction, the primary iron and steel industry, foundry work, logging/wood manufacturing, slaughtering, and agriculture. However, certain activities not generally considered as hazardous showed a high risk of repeat accidents (e.g., public hygiene, and the manufacturing of bricks, pottery, and glass).ConclusionsThe epidemiological pattern of repeat accidents by economic activity reflects that of accidents in general, though with exceptions. Results suggest that factors associated with specific technical aspects and production processes are important determinants of safety. The differences between single accidents and repeat accidents by gender, age class, and severity of injury suggest that studying repeat accidents separately from first accidents is quite useful.     

The critical re-action: Learning from accidents

How do organizations react and learn from accidents? The current paper deals with the justified necessity of acting upon, and learning from accidents. The aim is to discuss organizational dynamics and interplay, evoked and put into (re)action ‘after the fact’.The paper is part of an interview study that was designed to invite individuals to reflect back, and search for patterns in terms of how they interpret the learning impact of an accident, not just in terms of changes within their own organization that can be traced back to accidents, but for the sector as a whole. A presumption for this study was the importance to look at organizational interaction, i.e. how organizations act, affect, and speak with each other. In all 30 people from the rail and marine sector were interviewed with regard to (1) revealed changes in the aftermath of the accident, and (2) conditions for learning from accidents. Thematically, the interviews had two major accidents in Norway as a background theme: The high-speed craft MS Sleipner-accident (1999) and the railway accident at Åsta (2000).The paper reports and discusses key findings from the study, with a specific focus on the interplay between actors and parallel processes in the aftermaths of an accident.