Posted By David Bramall
Badger

Can I firstly ask a question - what is risk?

I think it is the probability of an event, i.e. the probability that a hazard will manifest itself in the form of an injury or illness to an individual or groups of individuals in very simple terms.

So we may ask what units can we use for the chance of throwing a dice and getting a 6 - to my simple mind there is no unit, it is a 1 in 6 chance. Thinking back to my schooldays, I can't remember any units involved in probability work, although I stand to be corrected if I have forgotten something.

I assume that you may use the standard type of risk assessment tool, 1,2,3, high, medium, low, or any other matrix which has become popular. If we go with the old "likelihood" 1,2 or 3, and "consequence" or "severity" 1,2 or 3, we could make a unit called the "LIKCO" or the "CONLIK" or something equally contrived, but it doen't really mean anything, does it?

Why not consider simply identifying the hazards and controlling them, do you really need a score or a unit to justify what you are doing?

Forget about risk - it doen't exist, if the hazard is controlled.

DrB

Posted By garyh
Quantified risk assessments I have used rank severity and probability numerically, with an associated risk when multiplied or combined. There are no units.

Posted By The toecap
It could be hits. No not hits on a web site

Posted By George Wedgwood
It's worth having a read of the HSE notes on; http://www.hse.gov.uk/risk/theory/alarpglance.htm
for the 'esxperts view' on risk calculation, which has moved on a wee bit since they published their groundbreaking Guidance book on "The tolerability of risk in nuclear power stations" That gave a very useful interpretation of risk, ALARP and how to calculate risk levels and probability, that still is very useful in hig hazard quantification work today. Once understood, it can still produce very persuasive arguments after calculation, about the right weight given to control measures.

Posted By Al..
Try this (it works for me).

The unit of measure is pounds per annum or euros per month or dollars per day. Whatever … it is “unit of money per unit of time”.

Risk = severity x likelihood

As everything can be reduced to money, severity can be measured in cash and we shall use pounds sterling (and measuring severity in cash terms is perhaps consistent with argument for the business benefits of good health and safety management.)

Likelihood is measured in “per unit of time”. For example, we might say that the likelihood is that this could happen is 2 times per year. We will use “per annum” as the unit of measure.

So risk is measured in “POUNDS PER ANNUM”.

(One way of handling risk is to take out insurance. Insurance underwriters are expert at estimating risk (or they don’t remain as underwriters for very long) The unit of measure of the insurance premium is pounds per annum so it all appears to hold together.)

Al

Posted By garyh
Using the much suggested Probaility x Severity you can produce a matrix - coloured in red, yellow or green squares. This is a very effective way to demonstrate the risk - if you are in the red (worse still, in the top corner!) - you need to worry.

I have seen this used (COMAH report) - Green is tolerable, Yellow requires formal Management review, Red requires immediate Management action to bring the risk into the "tolerable" region.

Seriously this model is very effective - if you rank your risks properly people really take note if you end up in the wrong sqaure.

Posted By the badger
Thanks to all for your comments but particularly to George, Al and Andy.

I had forgotten about the HSE's ALARP guidance which George recommends. I commend it to everyone as what formal risk assessment should really be about. High hazard industries where there truly needs to be some assessment of risk. Thanks George for that. It has set me on the right track.

I see that the HSE say on the ALARP pages "In a cost benefit analysis, we convert both risk and sacrifice to a common set of units – money – so that we can compare them." So risk is money - that was Al's point.

And I like Andy's PLL/yr - probable loss of life per year and the technique for factoring in injuries. Useful food for thought there.

Thanks again everyone.