Going right back to the original thread, you state the following formula as being universally accepted:

 risk = probability of accident × severity of its consequences

and then you go on to discuss "danger" and what is "danger" and how can it be classified.  

Well, in the UK we don't use the term "danger" and we don't start our risk assessments with the word "risk".  We start our risk assessments with the term "hazard".

The definition of "hazard" is "something with the inherent potential to cause harm or injury".  This might well be the same definition as your "danger" - something that is inherently dangerous.

We then look at the likelihood of harm or injury arising from that hazard - and that likelihood is the "risk".  When we consider the "risk" we look at who is likely to come into contact with the hazard, how often this is likely to happen and how severe any accident or incident would be.  We then look at the control hazards that we have in place and see how these protect the employee or environment from being damaged by this hazard.  

Once we have completed our machinations, we then give this a risk rating.  This is probably not in numerical terms, it may be high, medium or low risk, it may be colour coded, it may just be a warning statement.  

We then look at whether we can implement any further control measures to reduce the "risk" to the lowest level possible.  However the control measures need to be equal to the danger arising from the hazard.  You would not put a full extraction booth in place for a nuisance dust for example, and conversely, you would not rely on a "good airflow" for a carcinogen.  Everything needs to be worked out according to the risk posed and the benefits obtained.  This is in line with the UK legal term "so far as is reasonably practicable".

Once you have put your additional control measures in place, you can go back and risk assess again, taking into account the additional controls and, in theory, this will have reduced the risk arising from the hazard.  However, you also have to ensure that your control measure to limit one risk does not pose a risk in itself to another process.  For example, if you have an LEV to take away some fume from one area cut into a pre-existing system, you need to ensure that you are not causing this fume to come back down in a different part of the factory because the ductwork is not balanced properly.

That is how you complete a risk assessment in the UK.  I am quite sure everyone responding to you does it in pretty much the same way - however it looks on paper.  Whatever risk rating they use, everyone looks at the:

• hazard
• likelihood
• severity
• frequency
• control measures
• people

otherwise your risk assessment cannot be considered "suitable and sufficient" which is another UK legal term.

So now we come on to the accident.  If you look at any accident or incident investigation method you will find that there are multiple potential root causes, take taproot or tree diagrams, they're not named like this becuase there are only a handful of potential root causes, they are named for multiple causes, trunks, branches, twigs spreading outward from the centre.  Accident investigation is a serious subject in it's own right and quite complex.  The investigation from a RIDDOR accident can take weeks or months depending on external factors.  You cannot just look at the direct cause and go "yep, that's it" because the next person will have the same accident.  You need to trace this back to it's root and deal with the issue at that level, that way the accident will not recur.

I am quite sure you have your theories, but I think you need to go right back to the beginning (the root) and look at your formula.  Is this actually what people are using?  The answer in the UK is "no".  If you have built 20 years study on the wrong initial question then that is a very sad state of affairs and I feel very sorry for you.

Amungar

You wrote:

"For this reason, any theory of risk must "take the grain out of wheat and discard the straw", because if we do not we would be overwhelmed by a huge number of causes of a phenomenon. In mathematical terms we would have to consider thousands of causes of accidents and pathologies. Since this is crazy, it is necessary to consider only the "essential causes, or typical causes" that exist at all risk. All others must be discarded because they only exist in particular cases. To be able to do this, we must contemplate this matter from an abstract point of view, focusing on the commons that these causes have to discover what the primitive or basic causes really are, from which all the others derive."

Because most of us are assessing risk in the real world we have to consider thousands of causes and thousands of outcomes, together with the new risks that proposed solutions to one scenario might introduce.

At any given moment in time, we have to decide which potential risks are so negligible that we can discard the "straw". But circumstances change so what was straw in one year may become the source of more significant risk at a later date, or the reverse.

For example 2 years ago, I considered that the import of Chronic Wasting Disease into the U.K. from Scandinavia was a sufficient risk that we should take some precautions in some of our work to reduce the possibility that our staff might spread this disease with significant socio-economic consequences.

However, as  result of actions AND research taken by Norway and the EU, I can now reasonably conclude that the risks of what we do spreading CWD as a result of a Scandinavian source of danger (how you "define" danger or what this word "means" is frankly irrelevant) is negligible, so that we can stop certain precautions. This does NOT mean that we can also stop taking precautions to prevent our staff spreading the same disease as a result of a source in Northern America.

I can come to this conclusion with support from guidance from the relevant UK Ministry.

Now some of the precautions we would take would be the same whichever the geographical source, and some are precautions that we take to avoiding our staffing being responsible for the spread of other animal and plant diseases. 

At no point 2 years ago, nor more recently have I needed to apply some form of matrix type risk assessment.

Instead it's been about qualitatively assessing the risks based on evidence from numerous sources that includes the general presumption that were CWD to enter the UK, it would be almost impossible to eradicate it. That means that even applying the principles of "reasonable practicability" [that is explicitly included in much of our local legislation BUT which is implictly incorporated into EU law as then transposed into the law of many other EU Member States] demands a highly "precautionary approach". The probabilities of X, Y and Z become almost meaningless.

So we have to treat each scenario on its merits. We prescribe standards for e.g. the heights of barriers and rails to stop somebody falling from height, but these are based on an asssessment of MOST people at risk (predominantly based on an assessment of the UK population). As MOST people get taller then we have to re-evaluate national standards. 

So, I live in a house built in 1880, but converted into apartments in the 20th Century as have other houses in the same street. Each has a substantial drop from street footway level, so the street has cast iron railings. Where some of the conversions have included the addition of new stairs down to basement front doors, the railings are higher - people grew in average height, so the prescribed standard height to which the railings had to be constructed was increased.

Exactly the same applies to how we deal with e.g. protecting against flooding as we anticipate more severe consequences arising from climate change.

Edited by user 18 August 2019 13:18:42(UTC)  | Reason: Problem with how the fonts as typed translate when posted

Originally Posted by: amungar

If you or someone else asks me for more details I cannot give them to you because the explanation would generate even more details. The request for more and more details would never come to an end. Who wants to know more should buy my book, which I will edit next year. I intend to sell it also on Amazon. Its title will be: "Mathematical foundations of occupational risk engineering".

Wow, Dunning-Kruger at it's finest!!!

The Dunning Kruger Effect is a psychological phenomenon in which people of the lowest ability in a subject rate themselves as most competent, compared to others. Ironically, people who lack the most knowledge on a topic also lack the ability to recognize their own mistakes and errors, making them exceptionally confident and biased self-evaluators

I don't think this needs any further expansion......

Hilary - ouch!!

Marketing strategy might be less than optimal as well since (a) some of us try very hard not to buy from organisations that go out of their way to pay taxes in the country where the buyer is and (b) "occupational risk engineering" could be anything. Could be as simple as considering whether the design of a widget machine will actually work when manufactured, and then will people buy it.

...and even if the prospective buyer thinks about spending money, do I want to buy this book or read an article in an American paper on Lyme and other tick-borne diseases that I read this morning (for free)? Which would probably be better for my CPD? 

Hilary

Hi. In general, I agree with everything you say, but there are some concepts that I don't share and baffle me. For example:

1. You say: In the UK we don't use the term "danger" and we don't start our risk assessments with the word "risk". We begin our risk assessments with the term "hazard." I have seen that the Spanish translation of both English words is the same term "danger" (derived from Latin as the English word "peril" with the same meaning). However, from what you say later, I deduce that the word "hazard" is synonymous with the expression "harmful agent", because its meaning seems to be "source of harm." Instead, the meaning of the word "danger" seems to be "high probability of accident." Correct me if I'm wrong.

2. You say: Then we look at the probability of damage or injury arising from that danger, and that probability is the "risk." In my opinion, this is a huge leap, from the beginning of the canonical causal structure of risk (hazard) to the final effect of all causes (risk). All intermediate causes of risk are ignored. In my theory I highlight them by classifying them as "vulnerability", "threat", "incidentality". There can be no more classes: a) the vulnerability contemplates the space-time coexistence of an employee with a harmful agent / factor; b) the threat goes further because it contemplates the accessibility of an employee with a harmful agent; c) the incident goes beyond orque contemplates the physical contact between an employee and a harmful agent / factor. The primary cause vulnerability responds to your final comment, where you tell me "If you have developed the wrong initial question during 20 years of study, then that is a very sad state of affairs and I am very sorry for you." Now you know that the initial question (the initial cause), in addition to the “harmful agent” is the vulnerability of an employee to be in the same place and at the same time. Therefore, do not feel it for me because that is the correct initial cause in any risk theory.

3. You say: Accident investigation is a serious issue in itself and quite complex. In my theory, the canonical causal structure of risk is identical to the causal structure of any accident. The only difference is that the causal structure of the risk ignores the "casuistic" or "causal noise", while the causal structure of the accident does not ignore the particular cases of the specific case, which do not exist in all accidents of the same type, but Each one has its peculiarities. Therefore, it is not necessary to track the accident to its root, because the root of the risk and the accident is identical. Why? Because from a theoretical point of view the risk becomes an accident when there is physical contact of an employee with a harmful agent.

A cordial greeting

Sorry amunger, I have given you a perfectly reasonable and detailed explanation of how we do risk assessments int the UK and you have given me back paragraphs of what can only be described as gibberish.

I think your book will be a best seller - go for it!

Disagree all you like but I've had enough.

Lets just face it there are many ways of assessing risk, and one of those is by having a mathematical claculation which is conveniently placed on a risk matrix.  The size of the matrix will change the accuracy but not necessarly the outcome. So in theory a 5x5 matrix will give a different outcome to a 10x10 matrix.  However it is still down to a person deciding what number is allocated to the probability and the outcome so I may decide that a probability score for a process may be three and severity of the injury may be a four so the risk score is a twelve.  This will probably give a medium rating (work to do to reduce the risk).  The same process may see the scores doubled but in the end the process will still be a medium rating.

I guess I am saying that lets not get tied up with formula and their outcome.  Our profession would command a much higher level of respect if we kept it simple and easy for everyone to understand.

Let's recap the way this conversation topic started, which was to question the traditional risk assessment. At the insistence of several people asking me about the direct causes of risk, which I said are ignored by this formula, I was forced to leave this topic of conversation, which generated a lot of controversy with some of my ideas, which sadly were ridiculed by some people (maybe, because of insufficient information that interpreted some of my ideas absurd). Several people have written that this standard formula is used little or nothing. Strangely, they haven't told us what other alternative formulas they use. And we are eager to meet them.

In my opinion, the standard formula “magnitude of risk = probability of accident × severity of its consequences” is questionable for the following reasons (to avoid misunderstandings, there have been many, I have not said that this formula is wrong and I argue that nobody has the right to deprive myself of my right to question it):

1. This formula has been established dogmatically, which is why it is a conjecture. This formula is based on the authority of the one who published it for the first time, which, if I am not mistaken, was the US government agency under the Occupational Safety and Health Administration. Should we believe in this formula by the authority of its origin? The answer was given by Nobel Prize physicist Albert Einstein, in circumstances that may apply to this case: foolish faith in authority is the worst enemy of truth. Another answer comes from the philosophy of science: it tells us that any formula that does not rely on some theoretical model, but is invented based on the intuition or sagacity of its author, is a suspicious a priori formula because it is a conjecture while It is not proven. In this case, it cannot be demonstrated that this formula is wrong because of its dogmatic nature (it is based on faith). But it can be said that it is arbitrary because it lacks theoretical support, since it has not been deduced based on some accredited scientific theory or experimental facts.

2. It is a seriously incomplete formula because it ignores the pathology, which we know is the second type of possible harmful effect in a labor system.

3. If the value of the risk is different from zero, we accept the hypothesis that the probability of an accident is non-zero. This value implies that there is not enough technical or human security to avoid the accident, otherwise it would be impossible and its probability would be zero. But if there is not enough security to avoid the accident, what prevents it from happening now? The answer to this question is necessarily the following: while the accident does not happen, there is sufficient security to avoid it, which implies that the probability of an accident is zero during this period. But then a contradiction appears, because this probability is both zero and nonzero during the same period. One might think that, in the absence of sufficient security, what prevents the accident is that there is no technical or human failure that causes it. But this is a contradiction, because while there is no technical or human failure it is because there is enough security.

4. The mathematical concept accident probability considers a set of similar jobs and, consequently, with the same magnitude of the risk. This is an inconsistency, because we intend to determine the magnitude of the risk in a job chosen at random in a set, where we must know beforehand the magnitude of the risk in all of them, because they must be equal in all the jobs in the set. Otherwise, the probability of accident is meaningless. If the risk is not the same in each and every job, the calculation of the probability of accident is equivalent to calculating the probability of taking "six" in a set of dice loaded differently.

5. The formula is valid only in the case where safety measures are not applied that prevent harmful action during the incident of physical contact of an employee with a harmful agent (for example, does not consider the safety measure of an airbag , or of a differential electrical switch, which act while the harmful agent acts on the human body).

If anyone has anything to object to these considerations, I invite you to refute them with respect towards me, as well as my dealings with you.

A cordial greeting

Error correction

5. The formula only consider the case where safety measures are not applied that prevent harmful action during the incident of physical contact of an employee with a harmful agent (for example, does not consider the safety measure of an airbag , or of a differential electrical switch, which act while the harmful agent acts on the human body).

Originally Posted by: amungar

 Several people have written that this standard formula is used little or nothing. Strangely, they haven't told us what other alternative formulas they use.

Excuse me?  I wrote you down a formula for risk assessments.  It did not contain any numerical factors but, nonetheless, it was a formula.

Definition of formula:  "a method or procedure for achieving something".  

Edited by user 21 August 2019 15:00:07(UTC)  | Reason: Not specified

...have your ever listened to someone and then wondered who ties thier shoe laces for them?  :O

CptBeaky

Great answer as always.  Probably a waste of time but the rest of us I am sure enjoyed your reasoned response.

MARMALADE - I'm just going to put that out there because it's a big word that has no relevance to anything in this thread. 

Amunger

You are asking a group of people who do not use this formula and who have probably never seen this formula, whether it is complete or incomplete, has contradictions or is valid.  Answer:  we don't know and we don't care because we don't use it.  Therefore, we cannot either agree or disagree with your opinion as we have no working knowledge of this formula, no background to this formula, no working practices, procedures, standard work, legislation or codes of practice into which this formula fits.  It does not impact us in any way, shape or form.

The simplest answer to your original question of "is this formula correct" would be:

"we don't know because we don't use it and we cannot speculate on something that does not affect us". 

I said I'd stay away from this place, but the inner moth won't obey. I feel it's time for some real scientific arguments to support the current debate, and who better than the master of BS (i think this stands for bevahior safety? - D-Pak is King):

"On the periphery of official wisdom, if we can use that term to describe orthodox ways of thinking, millions of people have experienced and believe in the following propositions:

(Snipped list of woo beliefs)

What these diverse things have in common -- besides being excluded from official wisdom -- is that they imply an invisible connection.

By which I don't mean a mystical one. Just as the material world is connected invisibly at the quantum level, the subtle world is connected by a field of consciousness. A prayer, a desire to be healed, a wish for peace, hope for reassurance about the dead -- each impulse enters the field of consciousness and is responded to, just as every material event enters the quantum field and is responded to, down to the least quark and photon.

The first steps have to be taken, however, despite this welter of confusion. A culture of consciousness is possible. In fact, present-day science is such a culture, although it is based, ironically enough, on the premise that consciousness has no validity except as an emergent property of matter. One can foresee the next culture of consciousness based on connections, which would overturn the whole scientific prejudice against the subtle body, invisible realities, and the primacy of consciousness in general."

That basically sums up my view on things. What's your opinion on it as it pertains to the formula proposed? To me there's no question or doubt they go hand in hand.

Edited by user 24 August 2019 06:14:38(UTC)  | Reason: Reference

Dave

1. Can you pronounce that?

2. Can you translate it into Spanish?

Unfortunately when you translate

Arrrrggggghhhhhhhhhhhhhhhhhhhppppppppppphhhhhllluuuummmmmmmmmmp!

into Spanish and then back to English it ends up as the first twelve lines of the Magna Carta!  Who knew?

Edited by user 27 August 2019 07:35:56(UTC)  | Reason: Not specified