Risk is not any absolute value. Those matrices that people fill in are not objective measures (which is why the HSE does not particularly like them). It is entirely based on people’s  perceptions and even safety professionals will describe certain activities as “risky” rather than hazardous. In theory what we are trying to do is to reduce the risk so that all activities no matter how hazardous pose the same level of risk. In reality people often put more effort into high hazard activities while being blasé  about everyday hazards. There is that paper published decades ago which  compares the amount of money spent on risk control in certain sectors related to the number of deaths related to that sector. The nuclear industry came out on top with a low acceptance of risk, with millions spent to prevent each death while construction came out near the bottom.

Perception is everything: when an airliner crashes it is world  news  if some is killed in an RTA is barely makes the local news. Far more people are killed and injured in RTA’s around the world; they are one of the biggest killers of children.    

You can’t recalibrate all you can do is acknowledge this in yourself and talk to the stakeholders and try to establish what the organisations appetite for risk really is.   

Hi Self and Hasty

I think AK is referrring to HSE's "The tolerability of risk in nuclear power stations", but of equal relevance to your question might be "Reducing risks, protecting people".

I am going to argue that your current working environment is not nearly as low risk as you make out - it just doesn't have huge vessels with molten metal liable to be involved in an explosion and all the other things that can go wrong in a steelworks (including occupational health risks that have historically been poorly managed in the UK - I had a wake up call as part of an audit team in Ukraine!).

Now you may or may not be aware that there is a code of regulations which defines whether the enforcing authority should be the HSE or the local authority. There have been mulitiple iterations of that demarcation policy but some aspects have remained the same - including first and foremost that "higher risk" activities are dealt with by the HSE.

One key point is that if the "main activity" is wholesale or retail warehousing it falls to the local authority as being presumably "lower risk"

....and to be honest when I worked for HSE we didn't tend to spend much time at the warehouses apart from perhaps considering the structural integrity of racking systems and bulk storage of hazardous materials, and that was true whether it was a steelworks, bottling plant or bakery.

But this was a mistake as the statistics actually show that the warehouse is often one of the most dangerous parts of the entire operation! All those vehicle movements in the yard (and the ones on the roads outside that HSE generally chooses to avoid thinking about), accidents at taillifts and so on.

I have never been a fan of those numeric matrices that are so popular - almost invariable fiddled (deliberately or not!), both in terms of assessed severity and assessed likelihood of something going wrong, PARTICULARLY when calculating the "residual" risk if mitigations are in place - and even more so, if there is some directive that the "residual risk" MUST be determined as LOW (sometimes GREEN) or at worst MEDIUM (AMBER).

It is always possible that if you do the assessment properly some risks will still be HIGH (RED) even when all reasonably practicable measures are in place.

.....and if you use e.g. a 5 x 5 matrix it is VERY unlikely that it will adequately highlight the potential disaster that doesn't kill one or two, but many people. Unless you have a severity option of say 25, it is very easy for that type of incident scenario to entirely fall off the radar.

...and for that warehouse as an example one issue would be logistics. For some reason people seem to assume that a high speed vehicle impact kills and a low speed one doesn't - but Department of Transport statistics show that, if anything, the opposite applies!

While I agree to a great extent about about the use of numerical risk matrix tables, they have their pitfalls and but also their uses.

At least by trying to define and categorise various hazard/severity  and risk/frequencies of a potential scenario it does help to drive a level of consistency in determing (as other have said) the apparent risk priorities in an organisation. So yes in themselves the number mean little and somewhat subjective.

However, without such guidance - if you let 20 safety assessors loose  and they relied on their own gut feelings and experience you are likely to get a range of rik assesment result and hence prioities for action.

Accident statistics, I believe are still are a power full tool to consider, showing that slips, trips and falls on a level, falls from height and workplace traffic accidents - as Peter Gotch alluded too with his comments about working in warehouses.

But to answer the original question, do you recalibrate your views about workplace risks and hazards, yes - its called experience.

Hi Self and Hasty

Fully understand that you may have inherited a 5 x 5 matrix approach that people have become accustomed to.

So, if using such an approach the next task is perhaps to get all those using it to consider the variables - so as to get some sort of perspective of what you refer to as the percentage risk and avoid the extremes that people go to.

At one end of the spectrum they insist on considering the slips, trips and falls risks in the warehouse (as example) for EVERY task, so in effect if there are 10 tasks multiplying the risk ten fold.

At the other end of the spectrum, if you have 10 fork lift trucks operating in the warehouse, they do a risk assessment for "use of fork lift trucks" and so underestimate the volume of risk by the best part of a factor of 10 (which assumes that none of the FLTs are being used 100% in a shift).

...and at this end of the spectrum, they have a habit of lumping ALL the occupational health risks into one or two assessments, and so massively understate those risks.

As for the accident reporting, you probably need to look beyond the individual workplace to try and get a picture of the real risk - however well or badly internal reporting works.

...but even looking at the wider picture, underreporting means that some risks are more readily understood than others, particularly when it comes to the events that happen rarely but which are likely to be high severity.

Good luck, P

I wouldn't base it on either.  The most likely consequence is usually "nothing" but that doesn't mean you should do nothing!

I base it on the worst credible consequence (the worst thing that could reasonably be anticipated).

So if someone trips over, worst case is they land very badly and get a terrible head injury.  Most likely case is they aren't hurt at all.  But the worst credible consequence is a broken ankle.  So that's the severity I consider.

#10 Kate

Define 'credible' case is that the same as most likely.?  Whats the difference?

Whats the difference between 'most credible' and 'worst case'?

Playing Devil's advocate here!!

The difference is straightforward.

The statistics, if you had them, might be something like this:

90% of the time there is no injury except to the person's pride.

9% of the time there is a minor injury.

<1 % of the time there is a serious injury.

1 time in 100 million there is a fatality.

The most likely consequence is no injury as this has the greatest probability (90%).

The worst case is a fatality as this has the greatest severity (death).

The worst credible consequence is a serious injury as this is the worst thing that has a non-negligible probability (if something only happens one time in 100 million we choose to ignore it).

The principle is straightforward.

So is the practice.

#13 point 3

True there is no requirement to calculate risk, however the HSE R2P2 document does have numerical values as guidance, Hence if you are using statistics or other numerical data sources they have to be from a validated source to justify your ALARP decision

Without calculation you end up with a list of actions taken which maybe totally arguable if they reduce risk and by how much to achieve ALARP.