煤矿安全工作必须把人的生命安全摆在更加突出的位置而一刻也不能放松。为降低煤矿安全事故的发生率和对人的危害程度，要及时对煤矿安全程度进行风险识别、评估与预控。通过风险因素进行评估和预测控制并分析煤矿矿区的生产安全状况，对于煤矿实现“零事故”和安全生产以及保障员工的生命安全都具有重要的理论和实践价值。

本文首先对国内外矿工不安全行为相关研究进行统计与对比分析，并采用扎根理论的方法依据人因分析与分类系统理论（后简称HFACS理论）将矿工不安全行为风险因素划分为包含外部环境在内的五大因素，分别划分出16个构成要素及36个具体因素。其次，结合HFACS模型、模糊集理论和贝叶斯网络构建矿工不安全行为风险评估模型，从而细化影响因素指标并转换为模糊贝叶斯网络结构。然后，以模糊集理论为基础将该方法引入到贝叶斯网络结构分析中，专家对各影响因素指标进行评价打分，并将专家给出的定性自然语言变量转化为概率数值信息进行定量分析，将数据录入到模糊贝叶斯网络模型中并进行风险评估。最后，选取案例进行推理计算分析，得出该煤矿安全风险概率等级为中等，推理得出关键性的因素为：违反规章制度、精神疲劳、注意力不集中、未佩戴安全防护用具、未及时纠正违规违章行为、工作安排不科学不具体、生病服用药物或酗酒等六个指标存在较大的问题或故障；通过敏感性分析挖掘出该煤矿潜在隐患最大的四个因素分别为对员工安全和业务培训不到位、授权不具有资质的人员进行监督、事故处理不及时或措施不当以及认知情况和实际情况不一致。利用该方法进行预测和诊断分析的结果为矿工安全管理的决策提供了理论和技术支持，有助于提高煤矿矿区及矿工的安全性，实现安全生产。

基于人因分析与分类系统（HFACS）与模糊贝叶斯网络（FBN）的矿工不安全行为风险评估模型，解决了事件本身的不确定性并能够进行科学定量的概率推理分析，使计算结果更加符合矿区的实际情况也更具参考价值。与传统评估相比来看，对矿工不安全行为利用贝叶斯网络进行风险评估，能够同时兼顾实时性和实效性，为矿工不安全行为风险评估研究开辟了新思路，更提高了煤矿安全管理的风险控制水平。

Coal mine safety work must put human life safety in a more prominent position and cannot be relaxed for a moment. In order to reduce the incidence of coal mine safety accidents and the degree of harm to people, it is necessary to timely identify, evaluate and pre control the risk of coal mine safety. It is of great theoretical and practical value to evaluate, predict, control and analyze the production safety situation in coal mining areas through risk factors for realizing "zero accident" and safe production in coal mines and ensuring the life safety of employees.

Firstly, this paper conducts statistical and comparative analysis on the unsafe behaviors of domestic and foreign miners, and uses the method of root theory to divide the risk factors of miners' unsafe behavior into five external factors according to HFACS theory, which are divided into 16 components and 36 specific factors respectively. Secondly, the risk assessment model of miners' unsafe behavior is constructed by combining HFACS model, fuzzy set theory and Bayesian network, so as to refine the influencing factor indicators and transform them into fuzzy Bayesian network structure. Then, based on the fuzzy set theory, this method is introduced into the Bayesian network structure analysis. Experts evaluate and score the indicators of various influencing factors, convert the qualitative natural language variables given by experts into probability numerical information for quantitative analysis, and input the data into the fuzzy Bayesian network model for risk assessment. Finally, the case is selected for reasoning calculation and analysis, and it is concluded that the probability level of safety risk in the coal mine is medium. The key factors are: violation of rules and regulations, mental fatigue, inattention, failure to wear safety protective equipment, failure to correct violations in time, unscientific and unspecific work arrangement, taking drugs or drinking when ill; Through sensitivity analysis, it is found that the four biggest potential hidden dangers of the coal mine are inadequate safety and business training of employees, supervision of unqualified personnel, untimely accident handling or improper measures, and inconsistency between cognition and actual situation. The results of prediction and diagnosis analysis using this method provide theoretical and technical support for the management decision-making method of safety management of miners' unsafe behavior, which is helpful to improve the safety of coal mining areas and miners and realize safe production.

The risk assessment model of miners' unsafe behavior based on HFACS and FBN solves the uncertainty of the event itself and can carry out scientific and quantitative probability reasoning analysis, which makes the calculation results more in line with the actual situation of the mining area and have more reference value. Compared with the traditional assessment methods, the risk assessment of miners' unsafe behavior using Bayesian network method can give consideration to both real-time and effectiveness, open up new ideas for the risk assessment of miners' unsafe behavior, and improve the risk control level of coal mine safety management.

[1]国务院安委会制定部署安全生产十五条措施[N]. 新华每日电讯,2022-04-11(004).

[2]Groombridge C J,Kim Y,Maini A,et al.Stress and decision-making in resuscitation:systematic review[J].Resuscitation,2019,144(4):115-122.

[3]Tzampoglou P , Loupasakis C . Mining geohazards susceptibility and risk mapping: The case of the Amyntaio open-pit coal mine, West Macedonia, Greece[J]. Environmental Earth Sciences, 2017, 76(15):542.1-542.16.

[4]Zhang J , Xu K , You G , et al. Causation Analysis of Risk Coupling of Gas Explosion Accident in Chinese Underground Coal Mines[J]. Risk Analysis, 2019, 39(7):1634-1646.

[5]Maulana F A , Latief Y . Development of Risk-Based Standardized WBS (Work Breakdown Structure) for Safety Planning of Coal Mine Project with Surface Mining Methode[J]. IOP Conference Series: Materials Science and Engineering, 2020, 1007(1):47-63.

[6]Liu Q , Peng Y , Li Z , et al. Hazard identification methodology for underground coal mine risk management - Root-State Hazard Identification[J]. Resources Policy, 2021, 72(95):52-102.

[7]Jonek K I.Long-term analysis of the effects of production management in coal mining in Poland[J].Energies,2019,12(16):314-616.

[8]Palka D,STECULA K.Concept of technology assessment in coal mining[M].IOP Conference Series-Earth and Environmental Science,2019.

[9]Fa Z , Li X . From aggregate to part: Study on the change of coal mine safety records in China under the intervention of risk pre-control management system[J]. Resources Policy, 2021, 73(36):49-53.

[10]Timofeeva S S,Murzin M A.Man-made risks of coal mining enterprises.[M].IOP Conference Series-Earth and Environmental Science,2019.

[11]Shi T , Xinchun L I , Qiao W , et al. Effectiveness Evaluation of Coal Mine Risk Pre-control Management System Based on ANP-TOPSIS[J]. Safety in Coal Mines, 2019, 94(9):382-392.

[12]Bakhtavar E,Yousefi S.Assessment of workplace accident risks in underground collieries by integrating a multi-goal cause-and-effect analysis method with MCDM sensitivity analysis[J].Stochastic Environmental Research and Risk Assessment,2018,32(12):3317-3332.

[13]张俭让,史振东,闫振国,黄玉鑫,王智鹏,刘睿卿.煤矿安全双重预防信息管理系统设计研究[J].煤炭工程,2020,52(10):181-185.

[14]郜彤,张瑞新,郜赛超,何谦,刘贇.基于云模型和组合赋权的煤矿安全风险评价[J].工矿自动化,2019,45(12):23-28.

[15]王飞.煤炭企业系统性危险源辨识与风险评估研究[J].煤炭工程,2018,50(12):113-116.

[16]徐闯,何青松.基于多网融合及联动的煤矿安全监控系统设计[J].矿业安全与环保,2018,45(06):66-68.

[17]刘锦,武卫民.大型煤矿水灾害安全风险预警模型研究[J].中国煤炭,2017,43(05):122-124+129.

[18]Lawrence A C.Human error as a cause of accidents in gold mining[J].Journal of Safety Research,1974,16(2):121-124.

[19]Reason J.Human Error[M].U.K:Cambridge Universiti Press.1990,1-12.

[20]林泽炎,徐联仓.煤矿事故中人的失误及其原因分析[J].人类工效学,1996(02):17-20+70.

[21]傅贵,李宣东,李军.事故的共性原因及其行为科学预防策略[J].安全与环境学报,2005(01):80-83.

[22]曹庆仁,李爽,宋学锋.煤矿员工的“知-能-行”不安全行为模式研究[J].中国安全科学学报,2007(12):19-25.

[23]李琰,赵梓焱,田水承,于瑾慧.矿工不安全行为研究综述[J].中国安全生产科学技术,2016,12(08):47-54.

[24]满慎刚,李贤功,胡婷.基于中和技术的矿工不安全行为实证研究[J].中国矿业大学学报,2017,46(2):430-436.

[25]杨校毅,佟瑞鹏.基于Meta分析的矿工不安全行为JD-R模型[J].中国安全科学学报,2018,28(12):71-76.

[26]李广利, 严一知, 刘文椅,等. 基于DEMATEL-ISM的矿工不安全情绪形成因子研究[J]. 中国安全科学学报, 2021, 31(7):8.

[27]李磊, 李睿涵, 支梅,等. 基于非正式组织视角的煤矿工人不安全行为致因研究[J]. 煤矿安全, 2022, 53(1):5.

[28]Mohammadfam I,Ghasemi F,Kalatpour O,et al.Constructing a Bayesian network model for improving safety behavior of employees at workplaces[J].Applied Ergonomics,2017,58(11):35-47.

[29]Nie B,Huang X,Sun X,et al.Experimental study on physiological changes of people trapped in coal mine accidents[J].Safety Science,2016,88(38):33-43.

[30]Legault G,Clement A,Kenny G P,et al.Cognitive consequences of sleep deprivation,shiftwork,and heat exposure for underground miners[J].Applied Ergonomics,2017,58(27):144-150.

[31]满慎刚.矿工不安全行为分析及控制对策[J].煤矿安全,2013,44(12):217-220.

[32]王亚南,郭丽芳.控制力对矿工不安全行为的影响[J].煤矿安全,2018,49(05):248-252.

[33]李琰,乔立,于瑾慧.基于成本收益分析的矿工不安全行为影响关系研究[J].煤炭技术,2018,37(11):364-366.

[34]李琰,张燕.矿工工作压力对心理社会安全行为的影响机理及实证研究[J].中国安全生产科学技术,2019,15(03):135-140.

[35]李红霞,李妍雯.安全信息认知视角下矿工个体安全信息力影响因素研究[J].西安科技大学学报,2019,39(05):867-874.

[36]郭立君, 吕永卫. 工作家庭冲突对矿工不安全行为的影响研究[J]. 中国煤炭, 2022, 48(1):6-7.

[37]黄欣,李停军,邓增社,陈通,赵泓超.矿工不安全行为多层次模型及SD仿真研究[J].矿业安全与环保,2015,42(05):123-128.

[38]王晓波,栗继祖.基于ABC行为分析法的煤矿不安全行为HFACS模型[J].煤矿安全,2017,48(03):230-233.

[39]安景文,赵嘉丽,王金鹤,张冉冉.基于ANP-MEA模型的煤矿工人不安全行为评价[J].煤矿安全,2018,49(08):303-306+309.

[40]李琰,于瑾慧.矿工不安全行为成本和收益因素识别与分析:基于扎根理论的探索性研究[J].中国安全科学学报,2017,27(09):152-157.

[41]田水承,刘波,麻俊平.压抑情绪下矿工不安全情绪持续时间影响因素及干预对策研究[J].煤炭技术,2018,37(10):363-365.

[42]李琰,陈侠君.基于博弈视角的矿工不安全行为管理研究[J].煤炭技术,2022,41(03):231-234.

[43] Frey, Brendan J, Dueck, Detbert. Clustering by passing messages between data points[J].Science, 2017,315(5814):972-976.

[44] Heckerman D, Dan G, Chickering D M. Learning Bayesian Networks: The Combination of Knowledge and Statistical Data[J].Machine Learning, 2013, 20(3):197-243.

[45] Chickering D. M, Heckerman D, Meek C. A Bayesian approach to learning Bayesian networks with local structure[C].Thirteenth Conference on Uncertainty in Artificial Intelligence,2015.

[46] R. Chen, C. Chang. Using Bayesian networks to build data mining applications for a semiconductor cleaning process[J].International Journal of Materials & Product Technology, 2007, 30(4):386-407.

[47]A. Ouali, A. Cherif, M. Krebs. Data mining based Bayesian networks for best classification[J].Computational Statistics & Data Analysis, 2006, 51(2):1278-1292.

[48]王化琨, 王珍珍, 周影, 马维军.利用贝叶斯网络分析基因表达数据[J].数学的实践与认识,2010,40(8):70-73.

[49] Wu K, Ye S. An Information Security Threat Assessment Model based on Bayesian Network and OWA Operator[J]. Applied Mathematics & Information Sciences, 2014,8(2):833-838.

[50] Doguc O, Ramirez-Marquez J E. A generic method for estimating system reliability using Bayesian networks[J]. Reliability Engineering & System Safety,2017,94(2):542-550.

[51]耿文学,孔繁峨,马冬前.基于模糊贝叶斯网络的UAV中远距空战战术决策[J].火力与指挥控制,2014,39(03):99-103.

[52]李兴运,齐金平.基于模糊贝叶斯网络的受电弓系统可靠性分析[J].铁道科学与工程学报,2018,15(06):1383-1390.

[53]陈瑞,黄建文.基于模糊贝叶斯网络的工程项目进度完工概率分析[J].水电能源科学,2018,36(06):167-170.

[54]赵闯,郎坤.基于贝叶斯网络的生鲜物流风险评估[J].系统科学与数学,2020,40(11):2108-2124.

[55]尤亚云,李强,李金梅,李红钢.基于贝叶斯网络的消防协同治理影响因素研究[J].中国安全生产科学技术,2022,18(04):237-242.

[56]李新宏,朱红卫等.海底油气管道泄漏事故风险分析的贝叶斯动态模型[J].中国安全科学报报,2015,25(04):75-80.

[57]阴东玲,陈兆波等.煤矿作业人员不安全行为的影响因素分析[J].中国安全科学学报,2015,25(12):151-156.

[58]李贤功,葛家家等.煤矿顶板事故致因分析的贝叶斯网络研究[J].中国安全科学学报,2014,24(07):10-14.

[59]张宁,盛武.基于贝叶斯网络的煤矿瓦斯爆炸事故致因分析[J].工矿自动化,2019,45(07):53-58.

[60]张梅,许桃,孙辉煌,孟祥宇.基于模糊故障树和贝叶斯网络的矿井提升机故障诊断[J].工矿自动化,2020,46(11):1-5+45.

[61] Xu Ruihua,Luo Fan. Research on simulation of risk control strategy for air traffic controllers’ unsafe acts[J]. Safety Science,2022,151:78-105.

[62] Liou, James J. H.,Liu, Perry C. Y.,Luo, Shiaw Shyan,Lo, Huai Wei,Wu, Yu Zeng. A hybrid model integrating FMEA and HFACS to assess the risk of inter-city bus accidents[J]. Complex & Intelligent Systems,2022:1-20.

[63]白彦龙,李海立,陈昱,何舒蕾,白长江,西龙,薛玉壁,彭佳佳,王欣.基于HFACS的煤矿一般事故人因分析[J].煤矿安全,2021,52(05):250-255.

[64]安宇,李子琪等.基于FBN的高校实验室不安全行为风险评估[J].中国安全科学学报,2021,31(05):160-167.

[65]徐德宇,王天瑜,梁跃强,屈丽娜,李鹏.基于HFACS的化工企业火灾爆炸事故人因分析[J].中国安全生产科学技术,2020,16(11):66-70.

[66]林琳,郝立群,王赫,任丽辉.基于HFACS的粮食烘干作业危险源辨识研究[J].中国粮油学报,2019,34(S2):7-11.

[67]沈中芹,曾旺等.基于改进HFACS-MI模型的煤矿透水事故致因分析[J].安全与环境工程,2020,27(03):178-184.

[68]Liu R, Cheng W, Yu Y, et al. Human factors analysis of major coal mine accidents in China based on the HFACS-CM model and AHP method[J]. International journal of industrial ergonomics, 2018, 68: 270-279.

[69]柳茹林,程卫民,于岩斌,高天.基于HFACS-MI改进模型的煤矿重特大事故人因分析[J].煤矿安全,2017,48(08):250-253.

[70]王磊.基于改进HFACS的超高层施工事故人因关联性研究[J].智能城市,2020,6(23):64-65.

[71]王玉,王超,庄洁,宋润泽.基于改进HFACS的危化品罐车公路运输事故致因分析[J].物流技术,2019,38(09):68-73.

[72]庞兵,于雯宇.基于改进的HFACS和模糊理论的航空事故人因分析[J].安全与环境学报,2018,18(05):1886-1890.

[73]任智刚.新中国成立以来我国生产安全事故统计制度沿革分析[J].中国安全生产科学技术,2018,14(06):5-13.

[74]满慎刚,李贤功,胡婷.基于中和技术的矿工不安全行为实证研究[J].中国矿业大学学报,2017,46(02):430-436.

[75]田水承,郭彬彬等.煤矿井下作业人员的工作压力个体因素与不安全行为的关系[J].煤矿安全,2011, 42(9): 189-192.

[76]曹庆仁,宋学锋.煤矿员工的不安全行为及其管理途径[J].经济管理,2006(15):62-65.

[77]程恋军,仲维清.代际划分视角下矿工不安全行为研究[J].财经论丛,2016(01):90-96.

[78]任彦斌,陈文涛.煤矿本质安全风险管理理论研究[J].中国矿业,2008(04):41-43.

[79]何国家,杨春宁,黎体发.安全风险预控管理技术在我国煤矿生产实际中的理论方法研究[J].中国煤炭,2011,37(08):89-92.

[80]孟现飞,宋学锋,张炎治.煤矿风险预控连续统一体理论研究[J].中国安全科学学报,2011,21(08):90-94.

[81]桑志彪,彭锟等.基于风险管理理论的煤矿事故研究:人因风险因素探讨[J].中国煤炭,2014,40(09):121-124.

[82]曹庆贵. 企业风险管理与监控预警技术[M]. 煤炭工业出版社, 2006.

[83]孔繁琦.基于加权秩和比的安全生产风险评价方法研究——以新疆为例[J].中国安全生产科学技术,2013,9(04):55-59.

[84]刘双跃,江飞等.煤矿中人员角色与不安全行为的动态灰色关联分析[J].中国安全生产科学技术, 2013,9(3): 171-175.

[85]Judea, Pearl. Fusion, propagation, and structuring in belief networks[J]. Artificial Intelligence, 1986, 29(3): 241-288.

[86]肖秦琨, 高嵩. 贝叶斯网络在智能信息处理中的应用[M]. 国防工业出版社, 2012.

[87]许保光,王蓓蓓等.基于贝叶斯网络的航空安全中不安全信息分析[J].中国管理科学,2020,28(12):118-129.

[88]敦帅,陈强,丁玉.基于贝叶斯网络的创新策源能力影响机制研究[J].科学学研究,2021,39(10):1897-1907.

[89]高柯柯,于重重,晏臻.基于动态贝叶斯网络的智能工厂设备健康评估方法研究[J].机电工程,2021,38(06):768-773.

[90] Zadeh L A . Fuzzy sets[J]. Information & Control, 1965, 8(3):338-353.

[91]刘伟,刘晨君.改革开放四十年来煤炭行业安全发展之路[J].煤炭经济研究,2018,38(11):34-42.

[92]宋泽阳,任建伟等.煤矿安全管理体系缺失和不安全行为研究[J].中国安全科学学报,2011,21(11):128-135.

[93]黄智力,罗键.三角模糊数型不确定多指标决策的可能度关系法[J].控制与决策,2015,30(08):1365-1371.

[94]潘迎丽.基于改进BN网络的突发社会安全事件风险评估与预警研究[J].微型电脑应用,2021,37(12):48-51.

[95]黄晓家,王燚,谢水波,张雷.建筑火灾风险评估方法的发展与回顾[J].消防科学与技术,2021,40(08):1174-1177.

[96]姜传胜,曾明荣.突发事件分级与应急响应分级的关系辨析[J].中国行政管理,2021(02):131-136.

[97] 宋英华,马连隆,杨丽娇,蒋新宇.基于模糊BN的企业业务中断风险分析:以洪涝灾害情景为例[J].中国安全科学学报,2019,29(06):1-6.

[98] 肖琪聃,赵珍妮,刘林超.基于贝叶斯网络的地铁项目施工风险管理研究[J].信阳师范学院学报(自然科学版),2021,34(01):138-144.

[99] Li M, Wang D. Risk assessment of mine ignition sources using fuzzy Bayesian network[J]. Process Safety and Environmental Protection, 2019,28(6):33-38.

[100] 钟金花.踏石留印 抓铁有痕 推动安全生产专项整治三年行动落地见效[J].湖南安全与防灾,2020(07):8-8.