Risk Assessment of Decommissioning Options Using Bayesian Networks

Abstract

For complex and costly decommissioning activities, it is beneficial, if not necessary, that all relevant risks are identified and assessed on an overall basis, treating and assessing all risks within the same theoretical framework. Only then may the different options be consistently compared and the risks associated with decommissioning demonstrated and documented to the different parties of interest. The present paper suggests an approach for assessing the risks associated with the decommissioning of offshore facilities. The approach takes basis in a discrete point in time representation of the considered decommissioning options where important phases of the options are represented in terms of event scenarios. Using the possibilities of Bayesian Probabilistic Networks (BPN), the failure probabilities and risk events involved in the modeling of an option may then be analyzed for each phase and added up time-wise over the entire decommissioning process. The principles of BPNs are shortly described, and the proposed approach is illustrated by an example linking the operational and structural risks in connection with a re-float decommissioning option for a concrete offshore platform. It is shown how the sensitivity may be evaluated on the basis of the BPNs, thus providing a valuable framework to first improve the risk model in terms of the representation of important scenarios, then for deciding where to apply additional safety measures most effectively, and last but not least, to demonstrate and document the contributions to the mission failure probability.

Nuclear Decommissioning Risk Management Adopting a Comprehensive Artificial Intelligence Framework: An Applied Case in an Italian Site

“…The contribution of the relevant scientific literature is rather limited in presenting alternative NDP risk management processes and models to those suggested by IAEA. Some studies focus on the development of a qualitative assessment analysis of the technical and operational risks of a specific NDP being studied (Jeong et al, 2008), (Jeong et al, 2010), while others propose modelling alternatives for the identification and management of major risks through Bayesian Networks (Faber et al, 2002). Of particular interest is the contribution of Awodi et al (2021): their work focused on the identification of risk factors that are shared among many NDPs, focusing on how one or more risk factors may affect the success of the project.…”

Building biological indicators for soil quality assessment in Dong Thap Muoi (Mekong Delta Vietnam): An application of the expert Delphi

“…A number of methods have been suggested for building indicator systems, such as the Delphi technique (Dalkey, 1969), Bayesian approach (Faber, 2002), and multivariate analysis (Yazici, 2019). In particular, the expert Delphi technique, a structured expert elicitation process, is employed to identify a robust set of indicators.…”

Aftershock probabilistic seismic hazard analysis based on enhanced Bayesian network considering frequency information

“…More specifically, Bayesian networks (BNs) have been frequently employed in disaster risk analysis (Castillo et al., 2016; Hackl & Kohler, 2016; Zhang et al., 2023). In the area of seismic risk, Faber et al (2002) and Straub (2005) initially conducted pioneering research on the benefits of utilizing BN. Bayraktarli (2009), Bensi (2010), and Tien (2014) made contributions to the BN‐based modeling and the computational efficiency enhancement of civil and infrastructure systems.…”