Extending the dynamic and dependent tree theory (D2T2) to safety barriers: An application to an offshore fire deluge system

Abstract

Fault Tree (FT) Analysis is still one of the most common approaches for conducting reliability analysis of complex systems. However, it conceals several limitations such as the assumption of independent events and inability to model modern maintenance strategies in its basic form. To enhance FT analysis and overcome these limitations, Dynamic and Dependent Tree Theory, $D^2{T}^2$, has been recently introduced. This algorithm is based on the integration of Petri Nets, Markov models, and Binary Decision Diagram with the FT. Despite its effectiveness, its generalizability for different case studies needs to be explored. This is especially true for the application to safety barriers, which spend most of their time in a standby state. Thus, this paper aims to provide an extension of the $D^2{T}^2$ approach to further prove its capabilities and adopt it for safety barriers. To this end, the framework is applied to an offshore fire deluge system considering an availability analysis and a reliability analysis for the standby and operational phase respectively. The proposed modified framework could be used by maintenance engineers and managers to conduct reliability analysis of safety barriers, along with testing multiple maintenance strategies for their components.