(148c) 'qra Ain't Qra' - QRA Application in the Near Field - Our Experiences

Way back in the mid-1980’s when cinema audiences were treated to a time-traveling DeLorean vehicle in ‘Back to the Future’, HAZOP and Quantitative Risk Assessment (QRA) studies were starting their journey to becoming accepted tools for risk decision making in the United Kingdom and Australia.

Our experience has shown when conducting or reviewing Hazard and Operability (HAZOP) or PHA type studies, the assigned risk rating for operational related hazards may be overstated or in the worst case underestimated. In a HAZOP, team members may not appreciate the context of potential major accident type consequences (How Big) and impact (How Bad) to personnel and/or offsite population. Often these high consequence events are assigned an unlikely likelihood rating with a resultant low risk rating. There exists the potential to underestimate the risk.

As Trevor Kletz advocates, the conduct of a QRA provides a ‘rational and defensible method of deciding priorities’. A QRA will assist in identifying major risk contributors and assist in determining if the risk is tolerable and acceptable.

On Australian television, an advertisement for Castrol Oil had a memorable tag line ‘oils ain't oils!" where Castrol motor oil was shown to be of greater benefit to trouble-free engine performance than other oils. It is an apt metaphor which is central to the theme of this paper that ‘one size does not fit all’ or simply ‘QRAs ain’t QRAs’ when it comes to conducting risk assessments especially in the near field context.

Like a HAZOP, the QRA method must be applied with careful consideration of the required objectives and scope. This paper provides a case study on the challenges when conducting a QRA for a major petroleum storage terminal located in a heavily built up area with a mixture of industrial and residential land uses but also considered to be an aging asset.

By following ISO31000 Risk Management standard a QRA model (using commercial technical software) was built to allow the Operator to prioritize risk treatment options. By careful consideration of the study context, the QRA model accounted for the facility location relative to on-site occupied buildings and sensitive surrounding land uses, recognition of accidents past (eg Buncefield, Texas City), complementary safety studies findings (impact modelling, bowtie analyses, safety critical equipment and performance standards) to understand the risk profile.