Using dynamic Bayesian networks to model technical risk management efficiency

The objective of this paper is to present a mathematical model helping decision makers achieve optimum efficiency in risk management of product development. The optimum we are seeking considers qualitative data derived from expert opinions and quantitative information on project characteristics. The mathematical model proposed here aims at integrating data from these sources to identify opportunities for decreasing product risk. Reduction of overall product risk, before product release to production, is an indicator of the efficiency of the risk management intervention. Acceptable risk targets vary according to industry type, organization characteristics, regulations, etc. In general, the risk management process consists of identification of risks, analysis of risks, risk control, and feedback. Here, we propose a mathematical approach to risk management by using dynamic Bayesian networks for evaluation of product risks during the development period. The properties of the model are assessed by using 2 validation methods: k-fold cross validation and leave-one-out techniques. Mathematical imputation methods, like multivariate normal imputation, are invoked to deal with missing data. In addition, sensitivity analysis is performed to assess the uncertainty embedded in the parameters derived from the dynamic Bayesian network. Decision makers should consider the overall risk in product development estimated by this mathematical model. It may help to determine whether to release a product for Beta testing or to conduct additional activities to reduce the overall risk level before customer shipment. In addition, the model may be used for prediction purposes as it provides an estimate of the expected risk at time t + 1 based on the level of risk at time t.

[1]  Richard E. Neapolitan,et al.  Learning Bayesian networks , 2007, KDD '07.

[2]  Terje Aven,et al.  Models and model uncertainty in the context of risk analysis , 2003, Reliab. Eng. Syst. Saf..

[3]  David E. Booth,et al.  Analysis of Incomplete Multivariate Data , 2000, Technometrics.

[4]  Keiji Kanazawa,et al.  A model for reasoning about persistence and causation , 1989 .

[5]  Michael Luby,et al.  An Optimal Approximation Algorithm for Bayesian Inference , 1997, Artif. Intell..

[6]  Terje Aven,et al.  Risk assessment and risk management: Review of recent advances on their foundation , 2016, Eur. J. Oper. Res..

[7]  Changhe Yuan,et al.  Importance sampling algorithms for Bayesian networks: Principles and performance , 2006, Math. Comput. Model..

[8]  U. Lindemann.  Methodische Entwicklung technischer Produkte , 2009 .

[9]  J. Pearl.  TRYGVE HAAVELMO AND THE EMERGENCE OF CAUSAL CALCULUS , 2013, Econometric Theory.

[10]  Terje Aven,et al.  On when to base Event Trees and Fault Trees on Probability Models and Frequentist Probabilities in Quantitative Risk Assessments , 2012 .

[11]  David J. Spiegelhalter,et al.  Local computations with probabilities on graphical structures and their application to expert systems , 1990 .

[12]  Stuart J. Russell,et al.  Dynamic bayesian networks: representation, inference and learning , 2002 .

[13]  Nicola Torelli,et al.  Training and assessing classification rules with imbalanced data , 2012, Data Mining and Knowledge Discovery.

[14]  Elaine M. Hall,et al.  Managing Risk: Methods for Software Systems Development , 1998 .

[15]  Ron S. Kenett,et al.  Bayesian Networks in Survey Data: Robustness and Sensitivity Issues , 2016 .

[16]  Udo Lindemann,et al.  RISK MANAGEMENT IN PRODUCT DEVELOPMENT - CURRENT METHODS , 2006 .

[17]  Tyson R. Browning,et al.  Adding value in product development by creating information and reducing risk , 2002, IEEE Trans. Engineering Management.

[18]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[19]  John B Carlin,et al.  Multiple imputation for missing data: fully conditional specification versus multivariate normal imputation. , 2010, American journal of epidemiology.

[20]  Jonathan D. Cryer,et al.  Time Series Analysis , 1986 .

[21]  Carole Duval,et al.  Methodological developments for probabilistic risk analyses of socio-technical systems , 2009 .

[22]  Lakhmi C. Jain,et al.  Introduction to Bayesian Networks , 2008 .

[23]  Terje Aven,et al.  On the Need for Restricting the Probabilistic Analysis in Risk Assessments to Variability , 2010, Risk analysis : an official publication of the Society for Risk Analysis.

[24]  Enrique F. Castillo,et al.  Sensitivity analysis in discrete Bayesian networks , 1997, IEEE Trans. Syst. Man Cybern. Part A.

[25]  Ron S. Kenett.  Applications of Bayesian Networks , 2012, Trans. Mach. Learn. Data Min..

[26]  MenardiGiovanna,et al.  Training and assessing classification rules with imbalanced data , 2014 .

[27]  Trevor Hastie,et al.  An Introduction to Statistical Learning , 2013, Springer Texts in Statistics.

[28]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[29]  Alessandro Birolini.  Reliability Engineering: Theory and Practice , 1999 .

[30]  Guy M. Merritt.  Proactive Risk Management: Controlling Uncertainty in Product Development , 2002 .

[31]  Ron S. Kenett.  On generating high InfoQ with Bayesian networks , 2016 .

[32]  Kasturi Narasimhan,et al.  Reliability Engineering: Theory and Practice, 4th edition , 2005 .

[33]  Linda C. van der Gaag,et al.  Making Sensitivity Analysis Computationally Efficient , 2000, UAI.