Answers are grounded in this paper and cite their sources.
This paper presents the use of Bayesian networks as a methodology for eliciting and encoding Failure Modes and Effects Analysis (BN-FMEA) models of mechatronic systems. The method uses probabilistic directed acyclic graphs to construct causal dependency structures between functional, behavioral and structural random variables of the physical system. Default apriori probabilities and conditional probability tables are generated and attached to the graph structure for post-design evaluation by diagnostic engineers. BN-FMEA provides a language for design teams to articulate — with less ambiguity and greater precision — component failure cause-effect relationships across sub-systems. An example of an inkjet printer illustrates how BN-FMEA can be applied. The approach supports traditional FMEA objectives — identification of system failure modes, provides improved knowledge representation and inferencing power over spreadsheets, and is generally applicable to the class of industry- and government-standard FMEA spreadsheets and tables in use today. Finally, BN-FMEA is presented as a basis for improved integration of design and diagnostic modeling of mechatronic systems.
No takes yet. Share an insight, caveat, or question.
Burton Hoyt Lee (1999) studied this question.