I heard that the HSE does not like risk assessments that use number systems to evaluate the risks. Is this true I have always used the likelihood times severity method 1 being unlikely and 5 being near certain. Any thoughts?

We use the 1 -5 matrix and it suits us fine at the moment. It's certainly better than the method we used to used where we had a nonogram with likelihood on the left, severity in the middle and risk on the right with tie lines connecting them. Too complicated and wide open to abuse. General risk assessments are a very subjective process and trying to over quantify them is a fool's errand in my opinion. Unless the RA is for a very complicated process then the 5 by 5 matrix does the job for us.

The HSE/LA inspectors want to see hazards/risks controlled AFARP. How this is achieved is down to the individual company and will be determined by the complexity of the problems they face. There is no one size fits all solution, so if 1-5 works for you and the controls are suitable and sufficient that's that.

That's bizarre Claire, I've had positive experiences of HSE Inspectors to date, but evidently thay can be pretty arbitrary in their demands, John

Quote=lee said]I heard that the HSE does not like risk assessments that use number systems to evaluate the risks. Is this true I have always used the likelihood times severity method 1 being unlikely and 5 being near certain. Any thoughts?

I attended the HSE machine risk assessment course at Buxton a couple of years ago and while they didn't use a numerical system they did use a matrix that basically considered: Severity - Death, Permanent, Temporary, Slight How Likely - Very Likely, Likely, Unlikely, Remote and you ended up with a risk that was either - High, Medium, or Low. As other posts have said its more the conclusion you come to rather than how you get there is probably more important.

The problem with using numbers is in the interpretation as amply demonstrated in Larryl's post. For example, how can an accurate measurement or differentation really be made between unlikely and remote, likely or very likely, or temporary and slight? It is extemely subjective, that is made even more difficult when an organisation's assessments are are undertaken by a number of people. Also, what was 'very likely' in the morning could well be 'likely' in the afternoon, or vice-versa, etc as the dynamics of the activity/environment changes. It can also be tempting for the busy company assessor to fiddle the matrix numbers to give the (low) risk level desired, so the paperwork can be completed and they can then 'get on with the job.' Surely the more accurate and consistent approach is to ensure that whatever legislation and standards apply to the asessment/s being undertaken are identified and compared with the existing controls, with all non-conformances identifed and implemented as engineering and/or administrative actions, recorde in the further control measures column? I believe it is often overlooked what the MHSAWR says the purpose of risk assessment is - to make a suitable and sufficient risk assessment, for the purpose of identifying the measures he needs to take to comply with the requirements or prohibitions imposed upon him by or under the relevant statutory provisions. Could this be why so many employers find themselves on the wrong side of the law when their poor approach to risk assessment is judged as not being of a suitable and sufficient standard? Regards Peter