Process Safety and Risk Management using System Perspectives - A contribution to the chemical process and petroleum industry

Abstract

Risk management is all activities used to manage the risk of hazardous events and provides information to improve decision-making. A typical approach to system safety is to identify and eliminate the causes of an accident after it occurs and to repeat such efforts if a new accident occurs. (Ham, 2021). A traditional approach is principally reactive (Dallat et al., 2018). With the advancement of industrial systems, e.g., integrated control and safety systems, complex operation and shutdown sequences have evolved challenges in managing risk and safety. Considering the changing nature of today’s design and recent accidents, it has become vital to improve existing approaches to capture the complexity and dynamic nature of the automated system. The overall goal of the research presented in this thesis is to improve existing methods and develop new strategies using system engineering concepts and methodology for better risk management. Safety management focuses on two stages: pre-operational and operational stages. Design improvement of the system is focused on the first stage. The related tools can be utilized in the conceptual, preliminary, and detailed design stages. At the start of the design stage, detailed hazard identification should be conducted. The tools proposed for design improvement are inherent system safety and functional safety assessment. Safety assurance in the operational phase is achieved by monitoring safety performances. For monitoring safety, a performance indicator system-based perspective is advised. Based on the monitoring, safety training, education, regulatory compliance, inspection, or maintenance can be advanced, and plans can be set accordingly. The thesis is divided into two sections. Part I provides an overview of the risk management aspects to be considered. Part I also summarizes the main contributions of the research project. Some ideas for further work are also discussed in this part. Part II comprises six papers addressing different topics within the objective and scope of the thesis. The research focuses on various phases of the industry’s challenges in risk management. The thesis considers two hazard identification techniques: STPA and HAZOP, as hazard identification as the core of risk assessment in oil and gas activities. It questions whether present existing methods can identify hazards of the modern complex systems. It also proposes necessary improvements considering the complexity and interaction of today’s design. Present thesis discuss the topic of inherent safety and evaluation. Issues with the usage of inherent safety in the industry and practical challenges in adopting inherent safety indices by the industry and industry personnel are discussed. The pieces of literature discussed are relevant to the process and petroleum industry. It presents various inherent hazard risk factors with practical examples pertinent to the process industry. Identifying inherent hazards and risk factors makes it easier for the user to quickly find an inherently safer solution. The present research presents an inherent safety evaluation method for the system. The procedure is applied for a process system that validates the method’s applicability. The approach finds a scientific basis for previously established parameter-based inherent safety evaluation methods. The foremost step of the technique, which is finding inherent safety characteristics and their related parameter, makes the method flexible and general to be applicable in all industry sectors. The feature of a perfect, inherently safer system and their corresponding numerical values are determined to find a logical scoring system. The deviation of a real system for those parameters is determined to determine the score of inherent safety subindices; thereby overall inherent system safety index is determined. The method removes the problems of existing approaches, like dimensionality problems, lacking the logical basis of parameter scoring. The thesis also proposes a system engineering approach to check the adequacy of the facility’s safety barriers and safety assessment. Research adopts the FRAM (Functional Resonance Analysis Method) method to find the required safety barriers in the system. A two-level mathematical model is developed to predict the system’s safety. The developed method is applied with a practical case study of the Liquified Natural Gas (LNG) ship-to-ship transfer system. Furthermore, the thesis works on the development of safety performance indicators. It uses a system engineering method, System Theoretic Process Accident Model (STAMP), to develop indicators. Indicators were also developed using previously established methods like OECD (Environment, Health, and Safety Program) and CCPS (Center for chemical process safety). All the methods were applied for a case study of the LNG Floating Storage and Regasification Unit; Based on the evaluation, a comparative analysis was performed. The contributions of the research apply to several sectors and industry branches. Through the application of the methods, it has been possible to validate the developed methods and concepts. The thesis contributes to better decision support and improved risk management. The developed and analyzed methods focus on non-probabilistic methods. It emphasizes a non-probabilistic framework that does not depend on historical data. Assigning probabilistic information to an automated system is challenging and error-prone with excessive assumptions. However, the thesis points out the need to conduct more real case studies. Future publications should focus on applying the developed methods more straightforwardly to encourage users to use them. In addition, improved risk management methods should consider dynamic control of the automated system, which should also be focused on in future works.