In this article, we present a novel approach of modelling risk management process for complex systems. To overcome difficulties of modelling dynamic large-scale systems, the main idea is to split it into various structural homogeneous units. The object-oriented paradigm is used to this end but, unlike previous works, the proposed methodology allows variation in terms of internal parameters throughout the objects. This novel approach based on Bayesian network techniques is referred to as extended object-oriented Bayesian network. The main contribution of this article consists in establishing algorithms and methods on how to build and run such models. This article is an extension of a communication presented at AMEST by mainly developing a more realistic case study along with other improvements.
Bouzarour-AmokraneYTchanganiAPérèsF. Defining and measuring risk and opportunity in BOCR framework for decision analysis. (2012) In: 9e International Conference of Modeling and Simulation - MOSIM’12, 6–8 Jun 2012, Bordeaux, France.
CilliersP. Boundaries, hierarchies and networks in complex systems. Int J Innov Manag2001; 5(2): 135–147.
4.
Gell-MannM. What is complexity? In: CurzioAQFortisM (eds) Complexity and industrial clusters: dynamics and models in theory and practice. New York: Springer, 2002, pp.13–24.
5.
GoldenfeldNKadanoffLP. Simple lessons from complexity. Science1999; 284(5411): 87–89.
Hmelo-SilverCEPfefferMG. Comparing expert and novice understanding of a complex system from the perspective of structures, behaviors, and functions. Cognitive Sci2004; 28(1): 127–138.
8.
IkedaY. From complex system to integrative science. Diogenes2010; 57(3): 117–133.
9.
MainzerK. Thinking in complexity: the computational dynamics of matter, mind, and mankind. Berlin: Springer Science & Business Media, 2007.
10.
DuvalCFallet-FidryGIungBet al. A Bayesian network-based integrated risk analysis approach for industrial systems: application to heat sink system and prospects development. Proc IMechE, Part O: J Risk and Reliability2012; 226(5): 488–507.
11.
BromleyJ. Guidelines for the use of Bayesian networks as a participatory tool for water resource management. Wallingford: Centre for Ecology and Hydrology, 2005.
12.
JaegerM. On the complexity of inference about probabilistic relational models. Artif Intell2000; 117(2): 297–308.
13.
WeberPJouffeL. Complex system reliability modelling with dynamic object oriented Bayesian networks (DOOBN). Reliab Eng Syst Safe2006; 91(2): 149–162.
14.
BayesTPriceRCantonJ. An essay towards solving a problem in the doctrine of chances. London: C. Davis, Printer to the Royal Society, 1763.
15.
PearlJ. Probabilistic reasoning in intelligent systems. San Mateo, CA: Morgan Kaufmann Publishers, 1988.
16.
NielsenTDJensenFV. Bayesian networks and decision graphs. Berlin: Springer Science & Business Media, 2009.
17.
GodichaudMPérèsFTchanganiA. Optimising end-of-life system dismantling strategy. Int J Prod Res2012; 50(14): 3738–3754.
18.
GodichaudMTchanganiAPérèsFet al. Sustainable management of end-of-life systems. Prod Plan Control2012; 23(2–3): 216–236.
19.
HassenWBAuzanneauFPérèsFet al. A distributed diagnosis strategy using Bayesian network for complex wiring networks. IFAC Proc Vol2012; 45(31): 42–47.
20.
WeberPMedina-OlivaGSimonCet al. Overview on Bayesian networks applications for dependability, risk analysis and maintenance areas. Eng Appl Artif Intel2012; 25(4): 671–682.
21.
HassenWBAuzanneauFIncarboneLet al. OMTDR using BER estimation for ambiguities cancellation in ramified networks diagnosis. In: Proceedings of IEEE eighth international conference on intelligent sensors, sensor networks and information processing, Melbourne, VIC, Australia, 2–5 April 2013, pp.414–419. New York: IEEE.
22.
RabinerLJuangBH. An introduction to hidden Markov models. IEEE ASSP Mag1986; 3(1): 4–16.
23.
MurphyKP. Dynamic Bayesian networks: representation, inference and learning. PhD Thesis, University of California, Berkeley, 2002.
24.
FoulliaronJBouillautLAkninPet al. A dynamic Bayesian network approach for prognosis computations on discrete state systems. Proc IMechE, Part O: J Risk and Reliability2017; 231: 516–533.
25.
BangsoOWuilleminPH. Object oriented Bayesian networks a framework for topdown specification of large Bayesian networks and repetitive structures. Aalborg: Aalborg Universitetsforlag, 2000.
26.
KollerDPfefferA. Object-oriented Bayesian networks. In: Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence, Providence, RI, 1997, pp.302–313. San Mateo, CA: Morgan Kaufmann Publishers.
27.
LangsethHNielsenTD. Fusion of domain knowledge with data for structural learning in object oriented domains. J Mach Learn Res2003; 4: 339–368.
28.
WuilleminPHTortiL. Structured probabilistic inference. Int J Approx Reason2012; 53(7): 946–968.
29.
LiuQTchanganiAKamsu-FoguemBet al. Modelling a large scale system for risk assessment. In: Proceedings of international conference on industrial engineering and systems management (IESM), Seville, 21–23 October 2015, pp.203–208. New York: IEEE.
30.
LangsethHBangsoO. Parameter learning in object-oriented Bayesian networks. Ann Math Artif Intell2001; 32(1–4): 221–243.
31.
XiangYLesserV. Justifying multiply sectioned Bayesian networks. PhD Thesis, University of Massachusetts, Amherst, MA, 2000, pp.349–356.
32.
LiuQPérèsFTchanganiA. Object oriented Bayesian network for complex system risk assessment. IFAC: PapersOnLine2016; 49(28): 31–36.
