Red Room - Session 4
1:40 to 2:40 p.m. Wednesday, November 5, 2014
Common Cause Failure: Are you underestimating the probability of failure of your systems?
Common Cause Failure (CCF) reflects the simultaneous failure of two or more components due to a shared cause and is a critical consideration in system modelling and probabilistic risk assessment. Failure to model CCF will lead to underestimation of system risk and vulnerabilities in system design. Most statistical models assume independent failure events, which does not consider the dependencies of shared location, maintenance teams, manufacturing process deficiencies and other organizational and soft system design factors. This presentation will provide an introduction into the importance of CCF in modelling risk, and outline current methods to represent CCF in system models. We will identify the limitations of current techniques and, through examples, show how a new modelling approach – the General Dependency Model – overcomes some of these challenges.
Key Words: Common Cause Failure, Alpha Factor Model, General Dependency Model, Dependency Modelling, Probabilistic Risk Assessment
Andrew O’Connor and Benjamin Mailler