Using the emissions produced during the entire life-cycle of a fuel or a product, Life-cycle assessment (LCA) is an effective technique widely used to estimate environmental impacts. However, most of the conventional LCA methods consider the impacts of voluntary releases such as discharged toxic substances and overlook involuntary risks such as risk of accidents associated with exploration, production, storage, process and transportation. Involuntary risk of hazardous materials such as fuels could be quite significant and if ignored may result in inaccurate LCA. The present study aims to develop a methodology for accident risk-based life cycle assessment (ARBLCA) of fossil fuels by considering both the voluntary and involuntary risks. The application of the developed methodology is demonstrated for liquefied natural gas (LNG) and heavy fuel oil (HFO) as fuels of a hypothetical power plant. Adopting a Bayesian network approach, the comparative analysis of the fuels helps an analyst not only overcome data uncertainty but also to identify holistically greener and safer fuel options. (C) 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.