Common cause failures: definitions and typical errors

Abstract

The Aim of the paper is to analyse the definitions of the term “common cause failures” given in various international and Russian standards and point out their shortcomings; to identify and analyse the typical errors in the use of this notion and consideration of such failures as part of system dependability calculation. The importance of the topic is due to the fact that such failures reduce the efficiency of redundancy and must be taken into account in the process of the design of systems with high dependability requirements. Methods. The paper provides a comparative analysis of the definition of common cause failures given in Russian and international standards; analyses the methods of taking into account the effect of such failures presented in various publications; uses methods of the probability theory. Results. Differences between standards were identified in terms of the definition of the term “common cause failure”, as well as shortcomings of such definitions. Typical errors were pointed out in some publications dedicated to the methods of taking such failures into consideration. The simplest and most common beta-factor model was considered in most detail, the limits of its application were pointed out. Conclusions. It is advisable to use a single definition of common cause failures in different standards. It is to be taken from the basic terminological dependability standard with an appropriate reference. In the term itself and its definition, the word “failures” is to be in plural. The definition of this term in GOST EN 1070-2003 and GOST 34332.3-2021 are wrong, as they in no way correspond to the content of the defined notion. The conventional beta-factor model intended for taking into account common cause failures in the process of failure probability calculation can only be used in case of such probabilities being low.