Bayesian Networks Applications

 “Bayesian Networks are a powerful tool for knowledge representation and capturing in complex systems under uncertainties. The transparent structures of Bayesian Networks allow inferring roots of problems and influences of evidences on utilities and decisions features that facilitate the user acceptance and trust. Networks are the ideal amalgam of a data driven method and an expert driven method, allowing through large amount of data while still being able to explore the inherent relations and findings. Bayesian Networks have the ability to learn from observations. They include these findings either as changes in the network’s structure (which corresponds to changes in the cause – effect relationship) or as changes in the logic representations (which corresponds to changing the weight of different observations). “

Reference: Rosendahl, Tom Hepsø, Vidar. (2013). Integrated Operations in the Oil and Gas Industry – Sustainability and Capability Development – 16.2.2 Knowledge Capture Using Bayesian Networks. IGI Global.

Note: In the case of process safety BNs are an excellent medium to model complex cause-effect trees.

Technical Tools/Risk Assessment/Quantitative Risk Assessment/Frequency Techniques