Design for safety of engineering systems with multiple failure state variables

Since possible failure events of large engineering systems with a higher level of innovation may not be identified by experience or from previous accidents and incident reports of similar systems, and since ‘design for safety’ of such systems requires no omission of failure causes associated with possible system failure events, a top-down approach is not always satisfactorily applied in the risk identification and risk estimation phases and a more objective and flexible bottom-up approach may be more effective. aper proposes an inductive bottom-up risk identification and estimation methodology combining Failure Mode, Effects and Criticality Analysis (FMECA) and the Boolean Representation Method (BRM). This methodology can be used to identify all possible system failure events and associated causes, and to assess the probabilities of occurrence of them particularly in those cases where multiple state variables and feedback loops are involved. The Boolean representation method is presented together with its use in modelling cause and effect relationships. The overall model and the algorithms are described and tested in association with the associated computer software. The applications of this methodology in association with other formal safety modelling methods are discussed. An illustrative example is presented to demonstrate the methodology.