Development of a dynamic risk assessment framework for LPG transportation pipelines

Abstract

The present study introduces an innovative methodology for dynamic risk assessment of a hypothetical Liquid Petroleum Gas (LPG) offloading pipeline. The study mainly focuses on the determination of the probability of a catastrophic event dynamically, which is a major component in risk assessment. The output of this study is an open model for dynamic risk assessment of an LPG offloading pipeline with the potential of adopting it in any other application. The developed model presents the identification of the site and an analysis of the surrounding land uses, design, and related operations. Then it identifies the potential hazards. The traditional Bow-Tie diagram is created based on the identified risks and safety barriers. The Bow-Tie Diagram is then converted to a Bayesian network. The Bayesian network uses conditional probability tables which can be further improved for better reliability by introducing updated knowledge and experience. The method was trialled using a hypothetical scenario followed by a consequence analysis. A jet fire simulation is done using FLACS®, which is an industrial Computational Fluid Dynamics (CFD) code, to support the risk analysis. Financial losses connected with environmental damage, cleanup, evacuation, and lost output are among the consequences. The dynamic risk assessment framework presented in this study facilitates systematic decision-making on the LPG pipeline at almost any probable event. Further, it can be trained with experience and expert judgement