基于因果贝叶斯网络的风险建模与分析

工业工程

基于因果贝叶斯网络的风险建模与分析

1.北京航空航天大学经济管理学院，北京 100191； 2.华中科技大学经济学院，湖北武汉 430074； 3.北京振冲工程股份有限公司，北京 100102； 4.宾西法尼亚州立大学能源与矿产工程系，美国宾夕法尼亚州大学城，16802

出版日期:2016-10-31 发布日期:2017-02-21
作者简介:杨敏(1975-)，男，江西省人，副教授，主要研究方向为工程风险管理、专家判断抽取与集成、因果推断.
基金资助:
国家自然科学基金资助项目（71271014）

Risk Modeling and Analysis Based on Causal Bayesian Network: A Case Study

1. School of Economics and Management, Beihang University, Beijing 100191, China; 2. School of Economics, Huazhong University of Science and Technology, Wuhan 430074, China;3. Beijing Vibroflotation Engineering CO., LTD., Beijing 100102, China;4. Department of Energy and Mineral Engineering, Pennsylvania State University Park, PA, US, 16802

Online:2016-10-31 Published:2017-02-21

摘要/Abstract

摘要：

为验证基于因果贝叶斯网络的风险建模与分析（CBNbased RMA）的有效性，引入4种常见模式简化该方法的结构建模，以降低随后参数建模中专家判断工作量，然后将该改进方法应用于巴基斯坦NEELUMJHELUM水电站隧洞掘进工程风险分析中，有效控制了项目施工风险，获得远超预期的盈利。该案例应用结果表明，改进的CBNbased RMA方法具有很强的可操作性，可显著提高工程风险管理效率。

关键词: 工程风险管理, 因果贝叶斯网络, 隧洞工程

Abstract:

In order to validate the CBNbased RMA method proposed in article [1], four common patterns are first introduced to simplify the structure modeling of the method, which will reduce the effort of experts judgement in subsequent parameter modeling, and then the improved method is applied in risk analysis of NeelumJhelum hydropower station tunneling project in Pakistan. As a result, the contractors control effectively the project risks and obtain much more revenue than expected. This case shows that improved CBNbased RMA method has easy operability and will significantly enhance the efficiency of engineering risk management.

Key words: engineering risk management, causal Bayesian network, tunneling project

杨敏1，王竹葳2，王艳梅3，陈旭4. 基于因果贝叶斯网络的风险建模与分析[J]. 工业工程.

YANG Min1, WANG Zhuwei2, WANG Yanmei3, CHEN Xu4. Risk Modeling and Analysis Based on Causal Bayesian Network: A Case Study[J]. Industrial Engineering Journal .

参考文献

［1］PMI. A guide to the project management body of knowledge［S］. (Fifth Edition) USA: Project Management Institute, 2013.

［2］CONROW E H. Effective risk management: Some keys to success［M］. UK： Aiaa, 2003.

［3］ISO/IEC. 31010:2009Risk managementRisk assessment techniques［M］. USA: American National Standards Instit, 2009.

［4］COX L A. Risk analysis of complex and uncertain systems［M］. New York: Springer, 2009.

［5］HUBBARD D W. The failure of risk management: why it’s broken and how to fix it［M］. New Jersey: John Wiley & Sons, Inc., 2009.

［6］CABANO S L. Do we truly understand project risk?［J］. Engineering Management Review, IEEE, 2005, 33(1): 19-19.

［7］PEARL J. Causality: models, reasoning and inference［M］. 2nd ed. New York: Cambridge University Press, 2009.

［8］王双成, 唐海燕, 刘喜华. 用于风险管理的贝叶斯网络学习［J］. 控制与决策, 2007, 22(5): 569-572.

WANG Shuangcheng, TANG Haiyan, LIU Xihua. Learning bayesian networks in risk management［J］. Control and Decision, 2007,22(5),569-572.

［9］FENTON N, NEIL M. Risk assessment and decision analysis with Bayesian networks［M］. USA: CRC Press, 2012.

［10］O’HAGAN A, OAKLEY J E. Probability is perfect, but we can’t elicit it perfectly［J］. Reliability Engineering & System Safety, 2004, 85(1): 239-248.

［11］HORA S C. Eliciting Probabilities from Experts［M］//EDWARDS W, RALPH F. MILES J, WINTERFELDT D V. Advances in decision analysis: from foundations to applications. New York; Cambridge University Press, 2007: 129-153.

［12］王爱文, 杨敏, 段华蕾. 基于因果贝叶斯网络的风险建模与分析［J］. 系统工程与电子技术, 2013, 35(5): 1023-1030.

WANG Aiwen, YANG Min, DUAN Hualei. Risk modeling and analysis based on casual Bayesian network［J］. Systems Engineering and Electronics. 2013, 35(5): 1023-1030.

［13］PRELEC D. A Bayesian Truth Serum for Subjective Data［J］. Science, 2004, 306(5695): 462-466.

［14］CLEMEN R T, WINKLER R L. Combining probability distributions from experts in risk analysis［J］. Risk Analysis, 1999, 19(2): 187-203.