基于贝叶斯网络的航空安全中不安全信息分析

doi:10.16381/j.cnki.issn1003-207x.2020.12.012

Abstract

Abstract: In view of the unsafe events information collected in aviation safety, the data of the unsafe events found in the safety inspection is studied for the study. The problems found in the safety inspection are classified based on two classification principles, one of which is the factors related to data, and the other of which is related to human. The factors related to the human include training, supervision, staff deployment, information communication, knowledge, skills and several other aspects. The problem is divided into three categories:problems in the management of workflow, problems in the personnel management and problems in the equipment management. What's more, 10 basic reasons are summerized and the fault tree model of the civil aviation safety inspection is built. Besides, the fault tree model is used in the qualitative analysis of the minimum cut and structural importance. Then the NPC algorithm is used to infer the structure of the Bayesian network and study the parameters of the Bayesian network with the help of the EM algorithm. Combining the Bayesian network learning ability with the fault tree model, the Bayesian network model is presented for civil aviation safety inspection. Finally, based on the practical data collected from the airline security check, the impact of various factors is analyzed by use of the Bayesian network and come out with some conclusions. Individual factors, imperfect facilities and information communication are most likely to cause problems in the management of workflow. Management process and the deficiency of related knowledge and skills are most likely to cause problems in personnel management. Individual factors, training and equipment breakdowns are most likely to cause problems in device management. The research of this paper aims to provide an analytical method for the analysis and evaluation of aviation maintenance safety information. The Bayesian network model is constructed to solve the problem of civil aviation safety inspection in reality, which has pretty practical application value.

Key words: safety management, safety inspection, Bayesian network (BN), Fault Tree Analysis (FTA), influence factors

CLC Number:

F562

XU Bao-guang, WANG Bei-bei, CHI Hong, SHAO Xue-yan, GAO Min-gang. Insecurity Information Analysis in Civil Aviation Safety Based on Bayesian Network[J]. Chinese Journal of Management Science, 2020, 28(12): 118-129.

References

[1] 中国民用航空局. 2016年民航机场生产统计公报[Z]. 北京:中国民用航空局, 2017.
[2] 周志鹏,李启明,邓小鹏,等. 险兆事件管理系统在地铁施工安全管理中的应用[J]. 解放军理工大学学报自然科学版, 2009, 10(6):597-603.
[3] 王小兵,陈永青. 检查表法在安全评价中的应用[J]. 中国卫生工程学, 2006, 5(6):358-360.
[4] 中国民用航空局飞行标准司. 咨询通告:关于航空运营人安全管理体系的要求[Z]:AC-121/135-2008-26, 2008.
[5] Si Hu, Ji Hong, Zeng Xiaohong. Quantitative risk assessment model of hazardous chemicals leakage and application[J]. Safety Science, 2012, 50(7):1452-1461.
[6] Marais K B, Robichaud M R. Analysis of trends in aviation maintenance risk:An empirical approach[J]. Reliability Engineering & System Safety, 2012, 106(4):104-118.
[7] 王永刚,毛覃超. 基于模糊网络分析的民航机务安全评价[J]. 中国安全生产科学技术, 2012, 8(10):197-203.
[8] 刘洋,王燕青. 基于FAHP的航空维修安全风险管理研究[J]. 安全与环境学报, 2010, 10(1):189-192.
[9] 张元. 民航安全风险定量评价模型研究[J]. 中国安全科学学报, 2007, 17(9):140-145.
[10] 李亮,胡剑波,隋晓奎,等. 基于模糊理论的多属性航空安全风险评估[J]. 控制工程, 2007,(S2):33-35.
[11] Pearl J. Probabilistic reasoning in intelligent systems:Networks of plausible inference[J]. Computer Science Artificial Intelligence, 1988, 70(2):1022-1027.
[12] 张川,杨文雯,于超. 考虑风险传导情形的供应风险评估方法[J]. 运筹与管理, 2015, 24(4):172-177.
[13] 程华,李艳梅,罗谦,等. 基于C4.5决策树方法的到港航班延误预测问题研究[J]. 系统工程理论与实践, 2014,34(s1):239-247.
[14] 王超,金丽娜. 基于贝叶斯网络的进场航班排序队列稳定性研究[J]. 系统工程理论与实践, 2014, 34(1):262-267.
[15] 胡明礼,张彩芬,朱建军. 基于贝叶斯网络推理的双重不确定信息集结模型[J]. 中国管理科学, 2014, 22(2):135-141.
[16] Jun H B, Kim D. A bayesian network-based approach for fault analysis[J]. Expert Systems with Applications, 2017, 81:332-348.
[17] Luo Xiaochun, Li Heng, Dai Fei, et al. Hierarchical bayesian model of worker response to proximity warnings of construction safety hazards:Toward constant review of safety risk control measures[J]. Journal of Construction Engineering & Management, 2017, 143(6):04017006.
[18] Kim Y G, Lee S M, Seong P H. A methodology for a quantitative assessment of safety culture in npps based on bayesian networks[J]. Annals of Nuclear Energy, 2017, 102:23-36.
[19] Zarei E, Azadeh A, Khakzad N, et al. Dynamic safety assessment of natural gas stations using bayesian network[J]. Journal of Hazardous Materials, 2016, 321:830-840.
[20] Netjasov F, Janic M. A review of research on risk and safety modelling in civil aviation[J]. Journal of Air Transport Management, 2008, 14(4):213-220.
[21] Ale B J M, Bellamy L J, Roelen A L C, et al. Further development of a causal model for air transport safety (CATS):Building the mathematical heart[J]. Reliability Engineering & System Safety, 2009,94(9):1433-1441.
[22] Cook S C, Harris D. A civil aviation safety assessment model using a bayesian belief network (bbn)[J]. Aeronautical Journal-new Series-, 2005, 109(1101):557-568.
[23] Brooker P. Experts, bayesian belief networks, rare events and aviation risk estimates[J]. Safety Science, 2011, 49(8):1142-1155.
[24] 王广彦,马志军,胡起伟. 基于贝叶斯网络的故障树分析[J]. 系统工程理论与实践, 2004, 24(6):78-83.
[25] Jin Lijun, Peng Chenyi, Jiang Tao. System-level electric field exposure assessment by the fault tree analysis[J]. IEEE Transactions on Electromagnetic Compatibility, 2017, 59(4):1095-1102.
[26] 王长峰. 研发项目过程风险管理方法综合集成问题研究[J]. 中国管理科学, 2009, 17(4):148-155.
[27] Mao C, Canavero F. System-level vulnerability assessment for eme:From fault tree analysis to bayesian networks-part I:methodology framework[J]. IEEE Transactions on Electromagnetic Compatibility, 2016, 58(1):180-187.
[28] 史定华,王松瑞. 故障树分析技术方法和理论[M]. 北京:北京师范大学社, 1993.
[29] 中国民用航空总局. 民用航空器维修管理规范[Z]:MH/T 3010-2006, 2007.
[30] 中国民用航空局飞行标准司. 咨询通告:航空人员的维修差错管理[Z]:AC-145-15, 2009.
[31] 中国民用航空局飞行标准司. 维修单位的安全管理体系[Z]:AC-145-15, 2009.
[32] Shappell S A, Wiegmann D A. The human factors analysis and classification system-hfacs[J]. American Libraries, 2000, 1(1):20-46.
[33] Spirtes P, Glymour C, Scheines R. Causation, prediction, and search[J]. Lecture Notes in Statistics, 1993, 81(3):272-273.
[34] Steck H. Constraint-based structural learning in bayesian networks using finite data sets[D]. Institut für der Informatik der Technischen Universität München, PhD Thesis, 2001.
[35] Ren J, Jenkinson I, Wang J, et al. A methodology to model causal relationships on offshore safety assessment focusing on human and organizational factors.[J]. Journal of Safety Research, 2008, 39(1):87.

Insecurity Information Analysis in Civil Aviation Safety Based on Bayesian Network

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 2

Metrics

Comments

Recommended 10

[1]	REN Hui, WANG Dong-yu. Emergency Materials Reserve Strategies Considering Expected Redeployment Based on Cloud Services [J]. Chinese Journal of Management Science, 2020, 28(3): 31-39.
[2]	TAN Ying-shuang, ZHANG Xue-qing, HENG Ai-min. Study on the Scale and Influence Factors of Enterprise Production under Demand Uncertainty [J]. Chinese Journal of Management Science, 2018, 26(8): 138-145.