Comparer des méthodes

	Analyse des causes profondes bayésienne	Arbre d'événements (ETA)
Domaine	Plans d'expériences	Fiabilité
Famille	Process / pipeline	Process / pipeline
Année d'origine	1990s–2000s	2002
Auteur d'origine	Rooted in Pearl's Bayesian network theory (Judea Pearl, 1988); applied to RCA in process/reliability engineering from the 1990s onward	Andrews & Moss
Type	Probabilistic causal inference method	Forward inductive logic tree
Source fondatrice	Pourret, O., Naim, P., & Marcot, B. (Eds.). (2008). Bayesian Networks: A Practical Guide to Applications. Wiley. ISBN: 978-0470060308	Andrews, J. D., & Moss, T. R. (2002). Reliability and Risk Assessment (2nd ed.). Professional Engineering Publishing. ISBN: 978-1-86058-290-5
Alias	Bayesian RCA, Bayesian causal analysis, probabilistic root cause analysis, BN-RCA	ETA, Event Sequence Diagram Analysis, Initiating Event Analysis, Olay Ağacı Analizi
Apparentées	6	2
Résumé	Bayesian Root Cause Analysis (Bayesian RCA) integrates Bayesian network theory with structured root cause investigation to quantify the probability that each candidate cause is responsible for an observed failure or undesired event. Unlike deterministic RCA methods, it propagates uncertainty through the causal graph, updates beliefs as evidence accumulates, and ranks competing hypotheses by posterior probability — providing a principled, auditable basis for corrective action.	Event Tree Analysis (ETA) is a forward inductive technique used in reliability and risk engineering to model the possible outcomes that follow an initiating event. Starting from a single undesired event, ETA traces all subsequent event sequences through a binary branching tree representing the success or failure of safety barriers and protective systems. Introduced formally in reliability and risk literature by Andrews and Moss (2002), it is widely applied in nuclear, chemical, and aerospace industries to quantify accident sequence probabilities and guide safety decision-making.