Dynamic quantitative risk management of LNG bunkering SIMOPs

New international regulations aimed at decarbonising maritime transportation positively contribute to accelerating the use of liquefied natural gas (LNG) as a marine fuel. In recent years, the number of LNG-fueled ships has been rapidly growing, and as a result, LNG bunkering safety becomes an important prerequisite to ensure the safety of LNG-fueled ships. Fuel bunkering safety has always been a key issue of concern to the maritime industry because bunkering operations are potentially hazardous due to the high likelihood of leakage. Cryogenic and flammable LNG is a hazardous substance, posing new safety challenges during bunkering compared with conventional fuels. Furthermore, to maintain tight sailing schedules and to optimise the cost, it is suggested that LNG bunkering and the associated operations such as cargo loading/unloading, crew transfer, crane operations, and port activities be conducted simultaneously, which is referred to simultaneous operations (SIMOPs). SIMOPs pose additional risks to LNG bunkering because the operations are dynamically interlocked in which the occurrence probabilities of potential consequences change at different times due to the commencement or completion of specific SIMOP events. However, to date, there has been no detailed investigation and systematic research on the safety philosophy of and tailor-made risk management for LNG bunkering SIMOPs.

Given the above background, the overarching aim of this study is to improve the safety of the LNG bunkering SIMOPs through a better understanding of the risks and the development of a risk-based algorithm for quantifying and managing such risks. To this end, this thesis has two main objectives: the first is to develop a risk management framework for LNG bunkering SIMOPs; the second is to develop an approach to integrate safety philosophical factors (SPFs) into the risk management framework.

The synthesis of qualitative and quantitative methodological approaches is used to achieve the research objectives. The thesis's prominent qualitative research techniques are the document study, participation in activity, and case study. While numerous quantitative approaches, including online questionnaire survey, and several modelling techniques such as the Bowtie risk analysis model, the Bayesian networks, the cognitive reliability and error analysis method (CREAM), the ALOHA simulation, the calculation methods of probabilities of human death (POD), the fuzzy set theory, and the Onisawa function are employed to develop, demonstrate, validate and evaluate the quantitative risk management framework.

As a core part of the risk management framework, a novel dynamic quantitative risk analysis (DQRA) method is developed. In this method, the dynamic characteristics of LNG bunkering SIMOPs is captured to generate dynamic risk profiles, which can assist in making dynamic risk mitigation decisions timely. Five safety philosophical factors (SPFs) and the associated 23 SPFs indicators (SPFIs) are used to screen the safety performance of LNG bunkering. The SPFIs are integrated into the DQRA method through the human error probability (HEP) assessment that reflects the LNG bunkering industry's current and ever-changing safety levels. A new 'fuzzy BN-CREAM' is developed to assess the HEP. The integration with the SPFs and the DQRA provides a comprehensive dynamic quantitative risk management tool to enhance the safety of LNG bunkering SIMOPs. In order to make the DQRA method directly applied in the industry, a dataset for fuzzy occurrence probabilities of SIMOPs-related basic events is generated. The proposed method of generating the dataset can be used to continuously expand the database in the future. A case study was introduced to demonstrate, validate, and evaluate the developed risk management framework. The performances of five SPFs are investigated by ranking their SPFIs based on fuzzy crisp numbers. The result shows that safety management, safety training and safety culture are significantly insufficient in the current LNG bunkering SIMOPs.

To the best of the author's knowledge, there has been no systematic research in the literature specifically devoted to a study regarding the safety philosophy of, and risk management for, the LNG bunkering SIMOPs. Therefore, the development of a dynamic quantitative risk management framework is considered the main contribution to understanding the safety and managing the dynamic risks for LNG bunkering SIMOPs.