Why Hazard Studies Fail – and How to Get Them Right

Since being asked to write a column I have been very careful to not drift into my world of work. I work as a Process Safety consultant across multiple industries. My specialism is focused on the Functional Safety Lifecycle, including leading Hazard Studies amongst other work tasks.

I awaited Sean’s column in the June edition with his teaser image showing ‘Risk Consultants’ and some of the common issues that can occur. There is most certainly not ‘one-type fits all’ person who works in my industry.

I decided to drift into this topic as there have been some high-profile articles in the last month where hazard study inadequacies have been cited as a contributing cause of incidents.

Why can hazard studies contribute to incidents?

There are multiple reasons why a hazard study could contribute to incidents. The actual study itself doesn’t contribute to the incident but why could a hazard study be cited.

The reasons can include:

• Insufficient understanding of the system at the time of the study, either:
  • Design not mature enough
  • Personnel not fully understanding the system at the time of the study
• Insufficient time allocated for the study
  • We need to do it so do it quickly
  • Not allocating the correct resources
• Incorrect timing of the study
  • Too early – back to insufficient maturity
  • Too late – findings can’t be implemented

So, what does a good study NOT look like?

The following observations are not for any specific study, but are taken from many thousands of hours of study leadership. Success of a study is a balancing act, but the following can lead to unsuccessful studies.

• Having too many team members. I have experienced this on many projects where the room is full and there are on-line participants. You end up with one of two scenarios:
  • You never achieve a consensus of what the issues are hundreds / thousands of actions are raised and no one can see the true concerns that need addressing.
  • A small number of people drive the conversations and other team members are afraid of speaking up in front of such a large group and the future ramifications of their ‘stupidity’ during the session.
• Many changes of personnel. Different team members everyday / session (morning or afternoon). This can lead to the teams contradicting one another and then we end up with lots of rework and actions. Once again missing the real issues that need to be addressed.
• Using ‘Risk Ranking’ in studies where the risk matrix doesn’t seem right. Now risk ranking in hazard studies is a contentious issue at the best of time, but having a matrix that means every scenario evaluated is ‘Green risk’ and nothing is anything else is also a bad thing. You get a gut feel for what protections should be implemented (as do operators and vendor suppliers).
  • Vice-versa is also true where everything is ‘Bright Red’ and ‘Flashing like a Belisha beacon’ as this then means that the risk matrix is the opposite way.
• Incorrect number of parameter/guideword combinations for every node.
  • Too many (hundreds) can lead to too much ‘no concerns move on’ can lead to people turning off and issues being missed.
  • Too few can lead to insufficient discussion to prompt thoughts and missing concerns.
• Similarly incorrect node sizes
  • Too large and you can get confusion of where you are in the systems, participants can think everything is good when actually you are discussing the equipment that they have concerns with.
  • Too small and people lose concentration and people become ‘switched off’ similar to too many guidewords

How can you ensure that you have adequate hazard studies?

The key to hazard studies is to make sure that they are proportionate. Making sure that you have the right people in the studies for the correct amount of time with the right information.

The correct maturity of the design with the correct knowledge will lead to more successful results.

Unfortunately, good studies take time. Rush them and you might as well not bother.

Any changes completed post the hazard study do need to be re-reviewed to check that changes haven’t introduced new risks which could lead to new incidents / change existing ‘controlled’ scenarios becoming ‘uncontrolled’.

Make sure the rules of the study are understood and if the ‘belly’ is twitching that something isn’t right stop and discuss it. Everything being all good / bad can’t be right. Don’t get me wrong I’ve seen designs where I’ve said this should not move from paper and should definitely not be built! But this is unusual, when the study says it’s wrong but actually this is in line with good practice / what everyone else is building with the right protections then time to rethink the rules.

Regular review of your hazard studies whilst operating and gaining knowledge is essential for continued safe operation, those regulated under COMAH will be obliged to review studies for any changes every 5 years. But don’t forget changes in between these reviews. Even if it’s a quick review to confirm that the change doesn’t affect the previous hazard study evaluation will help maintain safety during operation of the facility. But you might find that a change solving one problem might give bigger headaches elsewhere!

Closing remarks…..

Make sure that your hazard studies are appropriate. Following proportionate rules and timings to ensure that the evaluation is useful and ensures that the correct measures are in place.

The incidents cited in the technical news didn’t blame the people but the processes. If you don’t know whether your processes are good enough you can review my points raised to see if you are guilty of any of these, or if really are struggling to know whether your processes provide the right level of assurance then reach out to someone for a review (look for that critical friend – I can be one of those if you can’t find one).

Trust me being a study chair you want to make sure that systems are implementing proportionate controls. We don’t have a scorecard on how many actions we raise with those raising the most getting a cuddly toy at the end of the year! I have been known to say some studies are ‘boring’ because the correct designs or controls are in place meaning that few actions are raised. Trust me, I sleep better at night knowing this is the case rather than actually telling someone that I don’t think this should ever be built!