Precursor-based hierarchical Bayesian approach for rare event frequency estimation: A case of oil spill accidents

摘要: Abstract Due to a scarcity of data, the estimate frequency rare event is consistently challenging problem in probabilistic risk assessment (PRA). However, use precursor data has been shown help obtaining more accurate estimates. Moreover, hyper-priors represent prior parameters hierarchical Bayesian approach (HBA) generates consistent results comparison conventional method. This study proposes framework that uses precursor-based HBA for estimation. The proposed method demonstrated using recent BP Deepwater Horizon accident Gulf Mexico. also applied same case study. show effective with regards following perspectives: (a) provides an opportunity take full advantage sparse available and add information from indirect but relevant data; (b) sensitive changes than method; (c) parameters, able model variability can exist among different sources data.