Automatic generation of a process shutdown system fault tree using a computer program KB3 (Sonatrach-FCP)

Automated dependability analysis has become a necessity due to the scale and complexity of modern systems. In this context, this paper is devoted to using a computer program called KB3 for probabilistic modeling of an operational process shutdown system's behavior, which ensures consistency, saves time, and is even easily accessible to non-specialists. Based on a graphical description of the structure and elementary characteristics of the system under study, using a generic knowledge base written in the Figaro language, the model developed automatically constructs the associated fault tree, which is then used quantitatively to obtain the probability of dangerous failure of the entire process shutdown system and its various constituent sub-barriers. This paper analyzes the performance of a process shutdown system using KB3 software, which is crucial for preserving operational well installations' safety. Initially using the reliability block diagram (RBD) knowledge base, the modeling process ensures a simple and flexible representation of the barrier system, including common causes of failure using the binomial failure rate model. The process produces a fault tree, which is time-consuming to build manually. Quantifying this fault tree gives the average probability of failure on demand (PFDavg) of the entire system and its components, which is essential for estimating the frequencies of occurrence of scenarios. However, to assess the impact of disregarding the existing standby relationship, the system is also modeled using the Boolean logic-driven Markov process (BDMP) knowledge base, showing no significant difference in numerical results.