Details

Autor(en) / Beteiligte

• Gerbec, Marko
• Aneziris, Olga

Titel

Uncertainties in failure rates in the LNG bunkering risk assessment

Ist Teil von

• Safety science, 2022-08, Vol.152, p.105774, Article 105774

Ort / Verlag

Amsterdam: Elsevier Ltd

Erscheinungsjahr

2022

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• •Risk from ship bunkering of LNG at ports is a hot topic in siting procedures.•LNG bunkering risk assessment studies and relevant data sources were analysed.•There is a number of comparability issues with the available data sources.•Failure rates of bunkering arms/hoses span two to four orders of magnitude.•Data sources and their use diverge and there is a need for a transparent data collection system. The increasing use of Liquefied Natural Gas (LNG) in shipping has led to an increased interest in marine bunkering safety in ports. The main tool for risk management and siting of bunkering facilities is quantitative risk assessment. This paper addresses the uncertainties in the available data on bunker equipment (loading and unloading arms and hoses), safety system failure rates, and ignition probabilities required for risk assessment. Using the search engines Scopus and Google, a literature search was conducted for studies and papers on the risk assessment of LPG bunkering. Analysis of the relevant studies revealed that different sources of data were used for equipment failure rates and ignition probabilities. The analysis of failure rates for leaks and ruptures of arms and hoses revealed that there are several problems with comparability, but also that there is a high degree of uncertainty, with data ranging from two to four orders of magnitude. The analysis of ignition probabilities showed that the uncertainty is about two orders of magnitude for small releases and within one order of magnitude for large releases. The failure rates of the emergency shutdown systems are within one order of magnitude. A wide range of arm and hose failure rates represents a large uncertainty for a reliable risk assessment. The way forward appears to be a transparent data collection system that describes the scope and the various assumptions.