A traceable process to develop Bayesian networks from scarce data and expert judgment: A human reliability analysis application

• The BBN quantifies errors of commission for probabilistic safety assessment We propose a novel process the quantification parameters allows traceable integration expert knowledge and (scarce) data Use to two sets some intermediate validation present paper develops Bayesian Belief Network (BBN) aggravating actions, as outcomes inappropriate decisions, be integrated in (PSA) models (i.e., so-called commission, EOCs). connects analyst ratings on influencing factors error forcing impact specific scenario, supporting CESA-Q method (the Quantification module Commission Error Search Assessment method). While contributing EOCs, this presents Conditional Probability Distributions, CPDs), striving data, form retrospective analyses operational events involving EOCs. combines functional interpolation populating CPDs updates adjust response available evidence. A first, prior is developed, then sequentially updated sets. This puts forwards steps future new EOC (or assessments) become available.