You've hit on one of the problems of over complicating the RA matrix, IMHO. Depending on the overall nature of the site "Bad" or 4 may mean 1 fatality whereas "Very Bad" or 5 may mean multiple deaths. So what does that make a 1? First Aid or less? and 2, a LTA with 3 a reportable?

For a low risk site 4 may mean a LTA whereas 5 could be a RLTA.

Then you have the likelihood to define. Again not easy unless you are talking specific failure modes.

And, in the final analysis, you'll probably end up with Low, Medium and High risks anyway.

Any particular reason why 5x5? I put a 3x3 matrix in place as I am always conscious of not putting staff off completing risk assessments. I work in social care so the complexity of risk assessments and staff culture may be different to your situation. At one time I considered removing risk ratings - in line with HSE sample risk assessments. However, I was pleasently surprised that staff wanted to keep them. In our case;

Severity; 'What's the worst harm you'd reasonable expect:

Slight = minor injury - up to three days off work
Serious = Can't work or do normal activities for over three days
Major = Permanent injury, illness or death

Likelihood =
Highly Unlikely
May Happen
Likely.

Results = Trivial, Low, Medium, High, Intolerable

Not perfect by any means. I don't think the rating is critically important in most cases - it's more the thought process that matters. We have risk ratings before and after additional control measures are put into practice. This encourages people to take a more determined look at how then can reduce risk - well that's the theory anyway!

I use one like this:

Likliehood:

1 - unlikely to happen
2 - possibly could happen
3 - likely to happen
4 - very likely to happen
5 - regularly occurs

Severity:

1 - slight injury, no lost time, no first aid
2 - minor injury - not reportable, but first aid treatment required
3 - more serious injury - over 3 day RIDDOR reportable
4 - Major injury - RIDDOR reportable
5 - Fatality

It's 5 x 6, but I'll throw it in anyway:


Severity of the hazard if incident occurs

1 - No Harm - No injury, damage, or sickness or other loss, i.e. of working equipment.

2 - Minor harm - First aid injury or illness, minor damage or loss.
For example, scratch, bruise, minor cut, minor burns. Normal work possible after first aid treatment.

3 - Serious harm - Burns, severe cuts, strains, sprains. Accidents result in loss of time less than 3 day injury.

4 - Major harm - Minor fractures (fingers & toes), temporarily disabling back injuries. Accidents result in an over ‘three-day’ injury or illness, substantial damage or loss.

5 - Catastrophic Harm - Fatality - Fatality, disabling illness, catastrophic damage or loss. For example, loss of limb, fractures to legs and arms, multiple injuries and fatalities.

_______________________________________________________

Likelihood of being exposed to risk

1 - Won’t Happen - The likelihood that a person will come into contact with dangerous machinery parts when there is fixed guarding in place.

2 - Extremely unlikely - The likelihood that a person will fall down a manhole with a properly fitted, intact cover. The likelihood that a person will touch a live conductor in a computer while word processing.

3 - Unlikely - The likelihood that a person will fall down an open manhole protected with suitable barriers and appropriate signage.
The likelihood that someone will touch a live conductor while servicing a computer which is switched off and isolated.

4 - Likely - The likelihood that a person will fall down an open manhole protected only by a flimsy barrier. The likelihood that someone will touch a live conductor while servicing a computer which is switched off but not isolated.

5 - Extremely likely - The likelihood that a person will fall down and open manhole which has inappropriate barriers or incorrect signage. The likelihood that someone will touch a live conductor while servicing.

6 - Almost Certain or imminent - The likelihood that a person will fall down an open manhole which has no barrier and no warning signs. The likelihood that someone will touch a live conductor while servicing a computer when it is switched on.

Reed described it well, however, we in the oil and gas have a breakdown of severity "People, Environment and Asset", as for people it starts with:
1 - slight injury, no lost time,
2 - minor injury - eg first aid treatment required
3 - more serious injury -
4 - Major injury including disability and a single fatality
5 - Multiple fatalities
Judging the frequency could be hard if you are part of a small organisation, but you can categorise as:
1 - unlikely to happen
2 - occurred else where
3 - occurred in the organisation
4 - occurs once a year in the activity
5 - regularly occurs in the same activity
You can define monitory valued for asset and impacts to the environment, but its all related to your line of business.

MEden

A martix is used to rank risks. If like some organisations you have hundreds if not thousands of real risks than a matrix is a good way of ranking them and dealing with them in order of severity.

How would I explain a severe accident with a 1/100 year rating to the next of kin, I would tell them I was dealing with the 1/1 year and 1/10 year accidents first as ranked by my matrix.

Thanks for all of yout comments, it seems to have prompted an interesting debate. My point in having a matrix was a means to scale and ranks the various risks that we are exposed to. Having done that I can deal with them in some kind of quantified order. I now have a 5x5 giving 4 levels of output, each having a different requirement:
1-4: OK
5-9: review controls
10-16 do not start until risks reduced to 9 or below
Above 17-25 (actually only 20 or 25 possible): stop process (this should only come up in a review)

So that starts part two of the debate..how do the rankings sound?

M - I like your approach, the rankings sound fine. I can't comment on the specific scales as they need to be right for your business and risk exposure but so seem on the right lines.

Ray - 'an idots guide' I'm afraid you're missing the point. A qualitive RA is fine for a small system with minimal risk exposure but when assesing comples high hazard systems then a matrix approach is the recognised industry standard. It in itself is a qualitative stystem with low ranking risks being assessed as just that. Depending on the type of risk then more detailed assessment or quantitative analysis is carried out on the higher risks. When you've managed all your higher risks you move onto the lower ones.

The HSE guidance is for small businesses not high risk organisaions who manage their risks properly, often using matrices in line with standards such as the Yellow Book, BS EN 50126-9 etc. Hardly a set of 'idiots guides'.

We are in a low risk environment ie Social Housing not petrochemical. We often identify several hundred hazards in a year and we have to budget carefully from our fixed income.

With the best will in the world we couldn't fix or control everything every year, that's including every potential trip point from a few mm upwards in public paths and so on.

So how can we manage these hundreds of risks?

In the past what could be repaired or controlled was doneuntil the budget allocation ran dry. Some were remembered for next years budget, many were forgotten until an accident happened or they were reported again.

We now use a 4X4 factor assessment to rank for budget priority, if that's an idiots guide so be it!

ahn2

I quite agree with you, for low risk organisations, complex risk matrix are not required (even the Sqn Ldr agrees with you).

I was hopefully helping people to understand how the risk matrix approach to risk assessment can be used to measure risk and to justify risk management actions etc.

Ray
'there are no right and wrongs, provided you can justify what you have done' - I think you should have added the caveat 'and a Court agrees with you'

HSE guidance may not be set in stone, but until told otherwise [by a Court], it remains a powerful/persuasive guide for the standard to be met.

Normally your answers are pretty good on this forum, but I think you were wide of the mark on this one. Just an observation - not a criticism

Just as an aside, whilst the HSE does not necessarily require a matrix form of risk assessment, they are not the only regulatory authority and at least one does require* a matrixed RA system.

*The ACoP actually says "This is a suggestion only" but who goes against an ACoP without VERY good reason!!

In order to interject some objective aspect into this subject, as safety practitioners we need to be aware of the research around the limitation of risk matrixes, particularly the research undertaken by Tony Cox, Louis titled ‘What's Wrong with Risk Matrices’?

In this he suggests “that risk matrices should be used with caution and only with careful explanations of embedded judgments”.

Dear all,

Before carrying out a risk assessment it should be remembered that we carry out assessments to identify the risks (severity and likelihood of injury) to the health and safety of employees and other persons arising from the work activity to identify the measures (needed to minimise, control and mitigate the effects of the risks) necessary to comply with the law. Furthermore, we have to put in place those measures, monitor them to see they are effective and keep records.

The matrix system implies that we only have to deal with significant risks; we do not - the law requires that we have to deal with all risks (that are real, and not fanciful or hypothetical risks) arising from the work activities of the undertaking. What is necessary is determined by the law: In some cases e.g. machine guarding we have to take specific measures regardless of cost; whilst in other cases we have to do what is practical, reasonably practical or what is reasonable in the circumstances.

Regards

akwatson statement

'The matrix system implies that we only have to deal with significant risks; we do not - the law requires that we have to deal with all risks (that are real, and not fanciful or hypothetical risks) arising from the work activities of the undertaking.'

All risks do not have to be assessed - formally assessing all risks is meaningless and overly restrictive/expensive exercise and is NOT required by the law.

Reg 3 Management Regs

(1) Every employer shall make a suitable and sufficient assessment of -
(a) the risks to health and safety of his employees

The ACOP goes on to say para 13 a
'Once the risks are assessed and taken in to account, insignificant risks can usually be ignored, as can risks arising from routine activities associated with life in general, unless the work activity compounds or significantly alters those risks'

So as well as looking at meaningful risks such as machinery safety, work at height etc do you have risk assessments for - making cups of tea, eating your lunch, taking stationary out of the cupboard, walking from your desk to a meeting room, turning the light on, coshh assessments for Tippex/liquids paper etc etc

a document that gives guidance on this subject BSI British Standards 18004:2008 Guide to achieving
effective occupational health and safety performance.
A large portion of the document is given over to Risk Assessment

Dear ITER,

Firstly. You must first assess a risk, otherwise how can you determine that it is trivial?

Secondly. I did not state that you had to deal with all risks; you could determine that it was not reasonable in the circumstances to do anything; that it was not reasonably practical to do more; or that it was not practical to do anything. However if somebody was injured then you would have to defend your position in court.

Furthermore when looking at risks; all risks have a probability spread of outcomes. If drink the tipex you may get cancer, liver damage, a headache or a raging thirst for more. So what do you put in your matrix, the most severe outcome and its probability or the most likely outcome and its outcome or both and their outcomes and then take the most serious rating.

I'm not an advocate of throwing the baby out with the bath-water, but i'd rather know the limitations of the tool. I'd also like to know that the tool does not protect the workers or the organisation. If this tool was a car it would have been withdrawn and banned.

Oh by the way making cups of tea scalding risk - severe outcomes to the young.
Unloading cupboards - manual handling - back injuries - significant person injury to old, overweight and pregnant.
Eating food - lead and other heavy metal poisoning in workshops.
Walking from your desk to a meeting room - slip, trips and falls - due to unsuitable floor surfaces, obstructions and poor lighting.

Regards