We have plenty of safety study methods available to help us understand risk and how it is controlled. HAZID, HAZOP, PHA, PHR, LOPA, QRA, SCTA, TRA etc. etc. They are all useful tools in the right hands but all have their limitations. Part of the skill is knowing which to use when, and how to interpret the results from each.
Ultimately the requirement is to reduce risk to As Low As Reasonably Practicable (ALARP). Although this idea has been around for a long time it is still poorly understood, including in a lot of the guidance that is available. It is easy to make it seem like a complicated issue that requires loads of data to quantify risk and the cost benefit analysis. In most cases it is actually very simple.
- What more can you do to reduce risk?
- Why are you not doing it?
You do need to have completed a suitable and sufficient assessment before you can claim that risks are ALARP. It’s not good enough simply to say you think it has been achieved. Also, you do need to continually review it, but that should not be confused with continuously adding more safety systems in some misguided approach tp continuous improvement.
At Hazards 34 (2024) I presented a paper suggesting the hierarchy of risk controls could be evolved to transform it from a high level concept to a useful tool. Here are the paper, presentation and proposed hierarchy (Excel)
https://abrisk.co.uk/wp-content/uploads/2024/11/2024-Haz-34-Hierarchy-of-Risk-Controls.pdf
https://abrisk.co.uk/wp-content/uploads/2024/11/Haz-34_030-Hierarchy-Brazier-01.pdf
https://abrisk.co.uk/abrisk-expanded-hierarchy-of-risk-control/
Nick Wise and I had a series of three papers published on this subject in The Chemical Engineer in 2001.
TCE April 2021 – What safety studies have you got on your menu?
TCE May 2021 – Risk tools of the trade.
TCE June 2021 – Making sure risks are ALARP.
Bow tie diagramss are a tool that can be use to illustrate risk controls and can have a role in demonstrating ALARP. But like all tools they have their limitations. Bow Tie Diagrams and Human Factors