Monitoring human and organizational factors influencing common-cause failures of safety-instrumented system during the operational phase

Abstract

Safety-instrumented systems (SISs) are important safety barriers in many technical systems in the process industry. Reliability requirements for SISs are specified as a safety integrity level (SIL) with reference to the standard IEC 61508. The SIS reliability is often threatened by common-cause failures (CCFs), and the beta-factor model is the most commonly used model for incorporating the effects of CCFs. In the design phase, the beta-factor, β, is determined by answering a set of questions that is given in part 6 of IEC 61508. During the operational phase, there are several factors that influence β, such that the actual β differs from what was predicted in the design phase, and therefore the required reliability may not be maintained. Among the factors influencing β in the operational phase are human and organizational factors (HOFs). A number of studies within industries that require highly reliable products have shown that HOFs have significant influence on CCFs and therefore on β in the operational phase, but this has been neglected in the process industry. HOFs are difficult to predict, and susceptible to be changed during the operational phase. Without proper management, changing HOFs may cause the SIS reliability to drift out of its required value. The aim of this article is to highlight the importance of HOFs in estimation of β for SISs, and also to propose a framework to follow the HOFs effects and to manage them such that the reliability requirement can be maintained.