Bayesian decision support for complex systems with many distributed experts

详细信息    查看全文

• 作者：Manuele Leonelli ; James Q. Smith
• 关键词：Bayesian decision theory ; Combination of expert judgement ; Decision support systems ; Graphical models ; Uncertainty handling
• 刊名：Annals of Operations Research
• 出版年：2015
• 出版时间：December 2015
• 年：2015
• 卷：235
• 期：1
• 页码：517-542
• 全文大小：1,294 KB
• 参考文献：Abbas, A. E. (2010). General decompositions of multiattribute utility functions with partial utility independence. Journal of MultiCriteria Decision Analysis, 17, 37鈥?9.CrossRef
  Ackermann, F. (1996). Participants鈥?perceptions on the role of facilitators using group decision support systems. Group Decision and Negotiation, 5, 93鈥?12.CrossRef
  Anacleto, O., & Queen, C. M. (2013). Dynamic chain graph models for high-dimensional time series. In Technical Report. Department of Statistics, Open University.
  Barclay, L. M., Smith, J. Q., Thwaites, P. A., & Nicholson, A. (2013). The dynamic chain event graph. In Technical Report. CRISM, The University of Warwick.
  Braziunas, D., & Boutilier, C. (2006). Preference elicitation and generalized additive utility. Proceedings of the National Conference on Artificial Intelligence, 21, 1573鈥?576.
  Cooke, R. M. (1991). Experts in uncertainty: Opinion and subjective probability in science. New York: Oxford University Press.
  Cowell, R. G., Dawid, P. A., Lauritzen, S. L., & Spiegelhalter, D. J. (1999). Probabilistic networks and experts systems. New York: Springer.
  Dawid, A. P. (1979). Conditional independence in statistical theory. Journal of the Royal Statistical Society. Series B (Methodological), 1鈥?1.
  De, S., & Faria, A. E. (2011). Dynamic spatial Bayesian models for radioactivity deposition. Journal of Time Series Analysis, 32, 607鈥?17.CrossRef
  Ehrhardt, J. (1997). The RODOS system: Decision support for nuclear off-site emergency management in Europe. Radiation Protection Dosimetry, 73, 35鈥?0.CrossRef
  Ehrhardt, J., Pasler-Sauer, J., Schule, O., Benz, G., Rafat, M., & Richter, J. (1993). Development of RODOS, a comprehensive decision support system for nuclear emergencies in Europe鈥攁n overview. Radiation Protection Dosimetry, 50, 195鈥?03.
  Faria, A. E., & Smith, J. Q. (1997). Conditionally externally Bayesian pooling operators in chain graphs. The Annals of Statistics, 25, 1740鈥?761.CrossRef
  Fishburn, P. C. (1967). Interdependence and additivity in multivariate, unidimensional expected utility theory. International Economic Review, 8, 335鈥?42.CrossRef
  French, S., Ranyard, D. C., & Smith, J. Q. (1995). Uncertainty in RODOS. In Technical report, School of Computer Science, University of Leeds.
  French, S. (1997). Uncertainty modelling, data assimilation and decision support for management of off-site nuclear emergencies. Radiation Protection Dosimetry, 73, 11鈥?5.CrossRef
  French, S. (2011). Aggregating expert judgement. Revista de la Real Academia de Ciencias Exactas, Fisicas y Naturales. Serie A. Matematicas, 105, 181鈥?06.CrossRef
  French, S., Maule, J., & Papamichail, N. (2009). Decision behaviour, analysis and support. New York: Cambridge University Press.CrossRef
  French, S., & Rios, Insua D. (2000). Statistical decision theory. London: Arnold.
  G贸mez M. (2004). Real-world applications of influence diagrams. In Advances in Bayesian networks (pp. 161-180). Berlin: Springer.
  Jensen, F. V., & Nielsen, T. D. (2013). Probabilistic decision graphs for optimization under uncertainty. Annals of Operations Research, 204, 223鈥?48.CrossRef
  Keeney, R. L., & Raiffa, H. (1993). Decision with multiple objectives: Preferences and value trade-offs. New York: Cambridge University Press.
  Lauritzen, S. L. (1992). Propagation of probabilities, means and variances in mixed graphical association models. Journal of the American Statistical Association, 87, 1098鈥?108.CrossRef
  Leonelli, M., & Smith, J. Q. (2013a). Dynamic uncertainty handling for coherent decision making in nuclear emergency response. In Proceedings of the winter meeting of the ANS, (to appear).
  Leonelli, M., & Smith, J. Q. (2013b). Using graphical models and multi-attribute utility theory for probabilistic uncertainty handling in large system, with application to the nuclear emergency management. In Data engineering workshops (ICDEW), pp. 181鈥?92.
  Leonelli, M., Riccomagno, E., & Smith, J. Q. (2015). Using computer algebra to symbolically evaluate discrete influence diagrams. In Technical Report (pp. 15鈥?7). CRISM, The University of Warwick.
  Little, R. J. A., & Rubin, D. B. (2014). Statistical analysis with missing data. Hoboken: Wiley.
  O鈥?Hagan, A. (2006). Bayesian analysis of computer code outputs: A tutorial. Reliability Engineering and Safety Systems, 91, 1290鈥?300.CrossRef
  Oliver, R. M., & Smith, J. Q. (1990). Influence diagrams, belief nets and decision analysis. New York: Wiley.
  Pearl, J. (1988). Probabilistic reasoning in intelligent systems: Networks of plausible inference. San Francisco: Morgan Kaufmann.
  Phillips, L. D. (1984). A theory of requisite decision models. Acta Psychologica, 56, 29鈥?8.CrossRef
  Queen, C. M., & Smith, J. Q. (1993). Multiregression dynamic models. Journal of the Royal Statistical Society, Series B, 55, 849鈥?70.
  Richter, K., Gering, F., & M眉ller, H. (2002). Data assimilation for assessing radioactive contamination of feed and foodstuffs. In Development and application of computer techniques to environmental studies, WIT Press, pp. 393鈥?02.
  Shafer, G. (1976). A mathematical theory of evidence. Princeton: Princeton University Press.
  Smith, J. Q., Barons, M. J., & Leonelli, M. (2015). Coherent frameworks for statistical inference serving integrating decision support systems. arXiv preprint arXiv:鈥?507.鈥?7394 .
  Smith, J. Q. (1989a). Influence diagrams for Bayesian decision analysis. European Journal of Operational Research, 40, 363鈥?76.CrossRef
  Smith, J. Q. (1989b). Influence diagrams for statistical modelling. The Annals of Statistics, 17, 654鈥?72.CrossRef
  Smith, J. Q. (1994). Plausible bayesian games. In Bayesian Statistics 5 (pp. 387鈥?06). Clarendon Press.
  Smith, J. Q. (2010). Bayesian decision analysis: Principles and practice. Cambridge: Cambridge University Press.CrossRef
  Smith, J. Q., & Anderson, P. E. (2008). Conditional independence and chain event graphs. Artificial Intelligence, 172, 42鈥?8.CrossRef
  Smith, J. Q., & Papamichail, N. (1999). Fast Bayes and the dynamic junction forest. Artifical Intelligence, 107, 99鈥?24.CrossRef
  Von Neumann, L. J., & Morgenstern, O. (1947). Theory of games and economic behavior. Princeton: Princeton University Press.
  Von Winterfeldt, D., & Edwards, W. (1986). Decision analysis and behavioral research. Cambridge: Cambridge University Press.
  Wakker, P. P. (2008). Explaining the characteristics of the power (CRRA) utility family. Health Economics, 17, 1329鈥?344.CrossRef
  Walley, P. (1996). Measures of uncertainty in expert systems. Artificial Intelligence, 83, 1鈥?8.CrossRef
  West, M., & Harrison, J. (1997). Bayesian forecasting and dynamic models. New York: Springer.
  Wisse, B., Bedford, T., & Quigley, J. (2008). Expert judgement combination using moment methods. Reliability Engineering and System Safety, 93, 675鈥?86.CrossRef
  Zheng, D., Leung, J. K. C., & Lee, B. (2009). Online update of model state and parameters of a Monte Carlo atmospheric dispersion model by using ensemble Kalman filter. Atmospheric Environment, 43, 2005鈥?011.CrossRef
  
• 作者单位：Manuele Leonelli (1)
  James Q. Smith (1)
  
  1. Department of Statistics, The University of Warwick, CV47AL, Coventry, UK
  
• 刊物类别：Business and Economics
• 刊物主题：Economics
  Operation Research and Decision Theory
  Combinatorics
  Theory of Computation
  
• 出版者：Springer Netherlands
• ISSN：1572-9338

文摘

Complex decision support systems often consist of component modules which, encoding the judgements of panels of domain experts, describe a particular sub-domain of the overall system. Ideally these modules need to be pasted together to provide a comprehensive picture of the whole process. The challenge of building such an integrated system is that, whilst the overall qualitative features are common knowledge to all, the explicit forecasts and their associated uncertainties are only expressed individually by each panel, resulting from its own analysis. The structure of the integrated system therefore needs to facilitate the coherent piecing together of these separate evaluations. If such a system is not available there is a serious danger that this might drive decision makers to incoherent and so indefensible policy choices. In this paper we develop a graphically based framework which embeds a set of conditions, consisting of the agreement usually made in practice of certain probability and utility models, that, if satisfied in a given context, are sufficient to ensure the composite system is truly coherent. Furthermore, we develop new message passing algorithms entailing the transmission of expected utility scores between the panels, that enable the uncertainties within each module to be fully accounted for in the evaluation of the available alternatives in these composite systems. Keywords Bayesian decision theory Combination of expert judgement Decision support systems Graphical models Uncertainty handling