The HSE doesn't even have L-M-H. http://www.hse.gov.uk/risk/casestudies/index.htm

Your second point is a little less black and white. It really depends on the nature of the industry (in my opinion). I am with you on this one though. Surely we (as professionals) already call on our knowledge of past incidences and known trends within the relavant industries when we do our risk assessments. This is what we are employed for! We don't work in a vaccuum. But it is much more important to look at your risk assessments individually. Going beyond differences in staff, you also have differences in workspace layout, lighting etc to deal with. What may be safe for one company could be deadly for another.

As for more than one person. It would depend on resources. Surely we already communicate with staff when carrying out risk assessments? Isn't this enough? Scoring isn't obligatory, so it is all about ensuring consultation with staff to identify the hazards.

So in summary. I think your thought process is correct, however. I use industry trends (from knowledge and experience) when conducting my assessments. I would never let them drive that assessment, just guide me in certain directions. I normally am more interested in the controls used within my industry than the hazards identified. (what sort of lifting equipment is used etc.)

Hi,

Check the auditor's background, he might be coming from process safety, with full QRA's, where often times statistical evidence is used.

Without such exterior information and statistics a risk matrix based RA will still be qualitative despite the "theory" as you'll use subjective knowledge and experience to come up with the numbers.

Personally I kind of hope some of the process safety rubs off on OHS, but I believe that's a long way away. Even in your current L-M-H you still use "What If" and similar process approaches, so I'm not sure why he would focus on a matrix based RA, perhaps a habit or a preference.

What's your line of work?  

I use agricultural worldwide statistics and try to keep up to date with industry trends. Chances are, that if people get crushed to death while driving ATV's in Australia and New Zealand (in alarming numbers), it might be a good idea to reconsider our farm transport.

I also keep an eye on MSDs and WRULDs in the food manufacturing industry and do regular ergonomics assesments including taking into consideration workforce differences (EU vs South Asia for example). This helps us tweak the height of workstations, positioning, twisting, etc.

You won't just use the stats, your collective input will still be valid. Just because dust explosions are rare, that doesn't mean we can't use the CSB's (Chemical Safety Board) videos and statistics to help us in our Planned preventive maintenance and housekeeping practices in Feed Mill.

Good advice from your assesor, kind of rare if I'm being honest. 

Good luck,

Mihai

Edited by user 04 September 2019 12:21:55(UTC)  | Reason: Engrish

Ignoring the matrix question as I believe there are enough posts already debating the relative merits - at the start of my employment the Paper and Board Industry Advisory Committee (PABIAC) was the go to source for published safety information in the days before the internet but it took a lot of industry commitment to support such activity.

Then if you look to the general notification of incidents there are often common themes e.g. baling equipment with interlocks faulty/over riden or guards removed - you may be the better employer  it will not stop components failing but if correctly executed should prevent contractors removing and not replacing guards https://www.shponline.co.uk/in-court/enva-toomebridge-limited-fined-after-workers-foot-trapped-for-45-minutes/ 

Edited by user 04 September 2019 13:20:16(UTC)  | Reason: FFS

We have a risk matrix dictated from head office that uses L-M-H for severity and L-M-H for likelihood with each weighted 1-2-3 and multipled together to give a rating.

I'm finding this a little odd since the form then states that anything scoring over 2 requires efforts to be made to reduce the risk, however for many things its going to be impossible to ever score less than 3. For instance a fall from a substantial height is always going to be a high in severity (3) and however unlikely it may be to occur the lowest likelihood score possible is 1 so whatever controls you put in place you'll always be compelled to look for more (and implement them within 6 months) according to the requirements of the form.....?

The L/M/H risk rating is reality an 'idiot guide' arguably for non-safety personnel, so they have a better grasp of hazards and the risks emanating from them. I think there might have been case some years back but now people who work in industry have a much better awareness of safety than say 20/30 years ago. Most are fully aware of the hazards and the potential risks which make the risk rating virtually redundant - it has become little more than a necessary legal requirement.  

As for insurance assessors...I'll leave that one for the time being.

Originally Posted by: ttxela

We have a risk matrix dictated from head office that uses L-M-H for severity and L-M-H for likelihood with each weighted 1-2-3 and multipled together to give a rating.

I'm finding this a little odd since the form then states that anything scoring over 2 requires efforts to be made to reduce the risk, however for many things its going to be impossible to ever score less than 3. For instance a fall from a substantial height is always going to be a high in severity (3) and however unlikely it may be to occur the lowest likelihood score possible is 1 so whatever controls you put in place you'll always be compelled to look for more (and implement them within 6 months) according to the requirements of the form.....?

I disagree. For example a fall arrest lanyard will not prevent you from falling (reduce the likelihood) but will prevent you from hitting the ground (reduce the severity).

My interpretation of what was said by the insurance assessor was that he was
not clear on your process for deciding H-M-L and suggested a matrix would at
least allow you to illustrate how you came up with your answer. In particular,
was the risk dominated by consequence or likelihood? Also, he was concerned
that you were discounting types of accident because you thought they were not
possible, but have occurred in industry. The "it can't happen here
because" attitude is very common and is often very strong until it does
happen at the company.

I am a bit bemused by the idea that L/M/H is only an ‘idiot guide.’ The
whole purpose of risk assessment is to focus your efforts on the highest risks,
which requires ranking. How you determine the ranking maybe an issue, but the
ranking is absolutely fundamental.

As for controls changing severity, this has been debated many times and 90%+
people agree they change the likelihood only.