Identification of Modeling Approaches to Support Common-Cause Failure Analysis
By:
Korsah, Kofi ; Wood, Richard T
Publication Date:
June 2015
Abstract
Experience with applying current guidance and practices for common-cause failure (CCF)
mitigation to digital instrumentation and control (I&C) systems has proven problematic, and the
regulatory environment has been unpredictable. The impact of CCF vulnerability is to inhibit I&C
modernization and, thereby, challenge the long-term sustainability of existing plants. For new plants and
advanced reactor concepts, the issue of CCF vulnerability for highly integrated digital I&C systems
imposes a design burden resulting in higher costs and increased complexity. The regulatory uncertainty
regarding which mitigation strategies are acceptable (e.g., what diversity is needed and how much is
sufficient) drives designers to adopt complicated, costly solutions devised for existing plants.
The conditions that constrain the transition to digital I&C technology by the U.S. nuclear industry
require crosscutting research to resolve uncertainty, demonstrate necessary characteristics, and establish
an objective basis for qualification of digital technology for usage in Nuclear Power Plant (NPP) I&C
applications. To fulfill this research need, Oak Ridge National Laboratory is conducting an investigation
into mitigation of CCF vulnerability for nuclear-qualified applications. The outcome of this research is
expected to contribute to a fundamentally sound, comprehensive technical basis for establishing the
qualification of digital technology for nuclear power applications.
This report documents the investigation of modeling approaches for representing failure of I&C
systems. Failure models are used when there is a need to analyze how the probability of success (or
failure) of a system depends on the success (or failure) of individual elements. If these failure models are
extensible to represent CCF, then they can be employed to support analysis of CCF vulnerabilities and
mitigation strategies. Specifically, the research findings documented in this report identify modeling
approaches that can be adapted to contribute to the basis for developing systematic methods, quantifiable
measures, and objective criteria for evaluating CCF vulnerabilities and mitigation strategies.