Risk is by nature closely related to hazard and, therefore, to uncertainty. Bayesian networks are commonly known to be well suited for risk assessment due to their ability to represent uncertain knowledge and to make rigorous probabilistic calculations.

Partners of the SKOOB project worked on risk and safety analysis applications for various socio-economic systems of strategic importance (nuclear, food industry, medical or social organizations). The above applications imply simultaneous integration of various dimensions (technology, organization, information, decision, and finance) correlated with the systems’ behaviours. This kind of application represents a good way to illustrate the power of PRM, as shown by some examples given on this site, for which the use of classical Bayesian networks would be too cumbersome or even impossible.

In a few words, we can say that PRM improve the process of engineering models, go beyond the modelling power of Bayesian networks, and help to use limited resources (hardware) for modelling and inference. Consequently, complex systems modelling becomes more tractable.