Hello colleagues:

I have doubts about the standard formula that assesses risk, which is even established in several laws as the only legal formula. My analysis is as follows and I await your opinion.

It is surprising that no current risk theory deduces the direct causes of risk. The universal standard formula risk = probability of accident × severity of its consequences does not define the direct causes of risk, but statistically assesses the risk in a high set of homogeneous jobs through its proven harmful effects. In other words, this formula values ​​the risk a posteriori, precisely when it has ceased to exist because it has become an accident. This formula could only be interpreted as a priori assessment of the risk in a given job if it is written as follows: risk = possibility of accident × amount of damage estimated in the specific case. It is necessary to eliminate the concept of accident probability in a given job because mathematics tells us that its value is constant ½, since there is only one favorable case (true accident) and two possible cases (true accident and false accident). It could be objected that this constant value of probability is not correct, because experience tells us that the presence of danger increases the probability of an accident. It is true, but then it is necessary to replace in the formula the concept of probability with the concept of danger. But this substitution causes an inconsistency because the concepts of danger and risk are similar. In addition, it is a seriously incomplete formula because it ignores the pathology, which we know is the second possible harmful effect of the risk. This analysis will baffle all the people convinced that this formula is correct, especially since it is the way in which the law obliges everyone to assess the risk (opinion of the top experts).

mathematical certainty relies upon physics which relies upon "fudge factors" a.k.a. constants those numerical factors whose absence prevent an equation from working

Thanks boblewis and Rondtuit for giving me your opinion. You're right in everything you say, but that's not the debate. The debate is the standard formula for assessing risk, which I think is inadequate because: a) it does not define the direct causes of risk in a specific case, with respect to a certain harmful agent (the evaluation must be repeated for each present harmful agent); b) does not assess the risk in a given job, but rather assesses the "statistical hope of damage" (product of the probability of the accident event by the amount of damage resulting) in a job chosen at random in a high set of similar jobs (with respect to a certain harmful agent, excluding others); c) does not assess the risk of pathology (with respect to a certain harmful agent, excluding others); d) Finally, this formula has no credibility because it has not been deduced from a current theory of risk generally accepted by the international community, but has been invented by some person, without any scientific basis. It is not, therefore, a scientific formula but a mere more or less ingenious recipe. 

What is your opinion?. Thanks in advance for your response. A greeting

Amungar

There are numerous "recommended" approaches, but in practice, the assessor should usually be well advised not to assume that the outcome (which we are trying to avoid) is the "worst case" which is often at least one death.

Statistics (e.g. collated in Great Britain by the Health and Safety Executive) indicate that for every person they know has e.g. fallen 5 metres, very few will die. Some will get broken bones, others other injuries.

....and the HSE only gets to hear about a fraction of the incidents involving people fall such distance, either because the incident does not result in a reportable event, or because reportable events such as "Specified Injury" or "over 7 day" accidents are massively under-reported.

Selecting the "worst case" when doing assessments using the method of calculation you have quoted will often result in scarce resources being diverted from where they are best applied.

But, as Ray has pointed out there are many other methods of doing risk assessment, some qualitative, some quantitative and some semi-quantitative. Sometimes past experience can help to inform these assessments, but ever more often we are dealing with new or changed circumstances where past history provides less assistance. So we then have to make judgements e.g. based on comparable scenarios or sometimes by applying e.g. the "precautionary approach" (which in itself may also draw on "comparable" scenarios - e.g. considering asbestos when assessing the risks from man-made mineral fibres of similar physical properties).

As Ray also says, sometimes translating theory into practice poses issues! 

I repeat the previous text because everything appears on the same line, being uncomfortable to read.

Pity. I strictly expected opinions on some considerations about alleged defects of the generally accepted formula for assessing occupational risk in general (for all cases without exception). But all the answers I have received are opinions about the difficulties that exist in assessing occupational risk, which has some peculiarities such as the uncertainty of the accident event. We all know this and my query has not raised these difficulties, but exclusively the variables that are used in that formula that traditionally values ​​occupational risk. Variables that do not contemplate "direct causes of risk" (which is what we are interested in knowing) but "direct effects of risk" (which we all know). Greetings

I think the point is ...can you answer my assignment question please...  :)

Thank you Peter Gotch for your kindness in answering my concerns so extensively.

You'll be right that I haven't explained myself well. For that reason I received answers that I did not expect from my question. Perhaps the reason is because my question is formulated in an abstract way, because it does not ask for an answer to solve a practical problem, but to solve a theoretical problem. And the people who go to these forums generally do so to ask for help to solve a practical problem in the exercise of their profession. For this reason the answers tend to give concrete examples, when what I request is an imaginary dialogue with a person interested in occupational hazards. The comments of my imaginary interlocutor are the kind of answers I expect. I think that will explain me better.

amungar: John, I have doubts about the effectiveness of the traditional formula for assessing occupational hazards in a given job: "risk = probability of (certain) accident x severity of its consequences". The reason is because the mathematical concept "probability" is applicable to a set of homogeneous events, where we want to know what the frequency of a given event is. Consequently, this formula is inapplicable to a given job because the frequency of the “accident” event is the constant ½, the result of the ratio between the favorable case “accident” and the two possible cases “true accident and false accident”. What do you think?

John: amungar, I think that for a given job the probability of accident should not be interpreted in a mathematical sense, but rather subjective, because the purpose is to estimate the ease of accident occurrence by applying intuition.

amungar: I understand you very well John, but then this formula should use the expression "accident facility" or, better yet, the expression "possibility of accident". But experience tells us that the magnitude of the possibility of an accident depends on the danger. In this case, if we substitute the imprecise concept “possibility of an accident” with the more precise concept “danger”, an inconsistency appears, because risk and danger are similar concepts, which now is not the case to go into detail to distinguish them. What do you think?

John: amungar, you have put me in a hurry because what you say is reasonable. I realize that the concept “danger” that, with very good criteria you have introduced, implicitly indicates that it would be necessary to replace the concepts “accident probability”, “possibility of accident” and “danger” by the concept of “insecurity in the work” because, at the end of the case, the presence of insecurity is the cause that facilitates the accident. Naturally, I do not mean any insecurity, but rather a certain insecurity that facilitates the occurrence of a certain accident.

amungar: John, I see that you have perfectly understood my doubt about that famous formula and, even, you have gone further because you have been able to understand the root of the problem. I agree with you that it would be convenient to replace the concept “probability of (certain) accident” with the concept “insecurity that facilitates the occurrence of a certain accident”. Congratulations!

What is your opinion of this imaginary conversation? 

NOTE: I apologize if there is any inconsistency in the text because it is an English translation using the Google translator.

Originally Posted by: amungar

Thanks boblewis and Rondtuit for giving me your opinion. You're right in everything you say, but that's not the debate. The debate is the standard formula for assessing risk, which I think is inadequate because: a) it does not define the direct causes of risk in a specific case, with respect to a certain harmful agent (the evaluation must be repeated for each present harmful agent); b) does not assess the risk in a given job, but rather assesses the "statistical hope of damage" (product of the probability of the accident event by the amount of damage resulting) in a job chosen at random in a high set of similar jobs (with respect to a certain harmful agent, excluding others); c) does not assess the risk of pathology (with respect to a certain harmful agent, excluding others); d) Finally, this formula has no credibility because it has not been deduced from a current theory of risk generally accepted by the international community, but has been invented by some person, without any scientific basis. It is not, therefore, a scientific formula but a mere more or less ingenious recipe. 

What is your opinion?. Thanks in advance for your response. A greeting

If you have not defined the cause of the risk, you have neglected the earlier step in the assessment process, put simply as 'who can be harmed, and how'. You need to consider the job, then the hazards and their causes, before tackling the calculation of severity and likelihood. The assessment needs to be tied in to a particular job, otherwise there could be no point in doing it. The 'risk of pathology' I'm not sure I understand. Contrary to what you say, I think you will find that the standard methodology is accepted internationally; an international standard is indeed quoted later in this thread.

I am going to be the person that says it.

Amungar the reason that you are not getting the responses that you expect is that you have phrased your question in such a long winded, complicated way that many of us are struggling to even understand what the questions is.

If what you are suggesting cannot be conveyed in simple language then how can you expect the general H&S advisor (or whatever their title is this week) to implement what you are suggesting? The risk matrix may not be without flaws, but it is a simple and very useful tool in helping people understand the concept of risk. It also is a very useful way of getting the lay person to look at hazards. Reducing a number is as straightforward as it gets when you want someone to understand a task is now "safer".

We can all agree that the risk matrix is not, in any way, a scientific formula. It is completely subjective what numbers go in. This is why most professionals don't even use the matrix anymore in doing their risk assessments. However to scrap it for something that you couldn't teach someone lacking in an understanding of higher level maths helps nobody. In fact it would just make H&S look even more elitist than it already does.

Let’s get back to the real world. There is (at least in the UK) no legal requirement to use the formal Risk = severity x likelihood. The HSE do not expect anybody to use this formula and at times HSE inspectors are quite hostile to it, is as people spend too much time looking at the risk assessment process rather than focusing on the important task of reducing the risk SFARP by the application of controls. Furthermore I think calling this a formula is a bit (ok very) ingenuous.  The numbers are not measurable values. For most people, most of the time, the severity and the likelihood are just picked out of the ether. Once you have numbers, it creates of false of security.

Hello to all the people who have been kind enough to give me their opinion on this thorny issue. I have followed the advice of "forestgate2016" and I have been overwhelmed by the amount of information available about the occupational risk assessment in ISO 12100 and in all the methods that are applied in the USA and in European countries. Of all that I have read (without deepening because it would take a long time to answer you), what has most caught my attention in the ISO 12100 norm considering that the “danger is a cause of risk”, because there are numerous schemes that put the danger first (exactly it says "detect sources of danger") to the risk estimate and then the risk assessment process. I have always believed that there is danger when, in the presence of risk, the probability of an accident is high, which prevents us from considering that the danger is a cause of risk. Another thing that has drawn my attention to the ISO 12100 standard is that it assesses the likelihood of an accident based on the frequency and duration of exposure (I suppose to the risk), completely ignoring insecurity in the workplace (as I mentioned in my previous analysis). The risk assessment in this standard does not contemplate at all the insecurity in the performance of a certain task. This baffles me, because if there is no insecurity the risk is zero. And experience tells us that the greater the insecurity, greater is the possibility of an accident. In other words, the probability of an accident depends on the previous existence of the possibility of an accident. 

I believe that we all realize that occupational risk is not yet a science but an art, where the best espert is the one who knows more and better safety measures to solve a specific problem. This reminds me of the way the ancient Egyptians worked to build. To project a right triangle they used a rope with equal spaces separated by three, four and five knots (the occupational risk considered as a recipe book to build safety measures). On the other hand, the ancient Greeks do not use recipes but science, specifically Pythagoras' theorem, with which they could build any right triangle (occupational risk considered as science, developing a theory endorsed by I'm afraid we will all have to wait for this new theory. In the same way that there is only one correct explanation for what has happened to a worker who has suffered an accident, regardless of whether or not one day we get to know it, we hope that there is also a single true version for the basic explanation of the operation of a labor system We live in the same universe, then we expect a single causal explanation of the facts. Paraphrasing what the subatomic particle physicist Witten said ironically (“If only one of the current five theories [of quantum physics] can correctly describe our universe and we affirm that the five are true, who then lives in the other four universes? ”): if only one of the current theories should explain the causality of the risk, but we affirm that all are true, which workers live in other worlds?” We will not know the answer to this question until all theories are discarded except one, which we accept is the only true one. As I believe, the reason we don't know the true answer is because existing theories correctly explain only some facts; they fail to predict other facts, or they cannot explain them. I don't have much more to say about this matter. Greetings to all.

I apologize for the Google translator

Originally Posted by: hilary

...you cannot measure risk in mathematic terms.......there will always be the "X" factor, that unknown quantity that varies from person to person depending on their soft skills such as attitude, perceptions, needs, motivation. 

You can measure risk in mathematical terms it's called actuarial science.  it essentially an fancy way of drawing up betting odd’s so essentially in the H&S world we are betting on how badly we think someone will get hurt…….. how ethical is that!

And the X factor….. fair enough but that focuses on the worker.  What about the context that the worker works in and all those things that shape the way the worker makes sense of their environment and influences the decisions they make (management, organizational culture, perception of priorities, goal conflicts etc, etc)

On the term “Probability”.  An event is either going to happen or it won’t in my opinion that’s relatively binary (0 or 1) there should be no level of probability there is nothing that will convince me otherwise.

Sorry guys I have been lurking in the shadows on this post but can’t contain myself any longer…the OP is clearly confused between what some people may be just semantics of the English langue and as they say it loses something in translation….

I would request that the OP look at research or further study on the difference between qualitative and quantitative processes….what is described here and what has obviously caused some confusion is qualitative…risk as we all know the combination of the severity and the likelihood that that harm would be realised, written in shorthand as likelihood x severity…it doesn’t imply that you have figures and multiply them it is just another method of expressing the theory…driving a car more often will increase to likelihood that you will be involved in an accident as an example…but it is subjective…quantitative methods (That Chris has alluded to) involved in the main peer reviewed scientific research or industry validated materials such as OREDA or HSE FRED and have mathematical formulas (TNO Yellow, Green, Red and Purple Books) which are also well practiced, along with those in industrial hygiene…so the English language can make like more difficult for all of us (English is not my first language – Scots Gaelic is) so I can understand the initial confusion…

https://www.ncbi.nlm.nih.gov/books/NBK218582/

Sorry guys I have been lurking in the shadows on this post but can’t contain myself any longer…the OP is clearly confused between what some people may be just semantics of the English langue and as they say it loses something in translation….

I would request that the OP look at research or further study on the difference between qualitative and quantitative processes….what is described here and what has obviously caused some confusion is qualitative…risk as we all know the combination of the severity and the likelihood that that harm would be realised, written in shorthand as likelihood x severity…it doesn’t imply that you have figures and multiply them it is just another method of expressing the theory…driving a car more often will increase to likelihood that you will be involved in an accident as an example…but it is subjective…quantitative methods (That Chris has alluded to) involved in the main peer reviewed scientific research or industry validated materials such as OREDA or HSE FRED and have mathematical formulas (TNO Yellow, Green, Red and Purple Books) which are also well practiced, along with those in industrial hygiene…so the English language can make like more difficult for all of us (English is not my first language – Scots Gaelic is) so I can understand the initial confusion…

https://www.ncbi.nlm.nih.gov/books/NBK218582/

Originally Posted by: O'Donnell54548

Amunger - have you run out of fluff to pull out of your navel and now need something else to busy yourself with....?????? Your attempts at demonstrating your superior intelligence has failed.

Although there is a saying that says "foolish words, deaf ears", this time I will make an exception. You are right about one thing: if all that has occurred to you to answer my doubts about the traditional formula that values ​​risk is to make fun of me, then it is true that my intelligence is superior… ..to yours (and of all who have declared that they approve your answer). Your foolish words have not hurt me. What hurt me is that no one else has censored your foolish words, looking the other way. As history has always told us that "the wise are not always right", I do not think that questioning what the wise say about occupational hazards has bothered you, those who have supported you and those who have kept silent by putting themselves in profile. I suspect that what has bothered you is to have said that you still don't have a “risk science” but a “recipe book” (manual of safety measures) to solve your problems in the exercise of your profession. But the proof is evident: when the security measure you need is not written in the recipe book, then you do not go to science to enlighten you, but you go to forums like this so that someone more cunning or wiser tells you what is the best recipe to solve that particular problem. This is because your knowledge is not yet worthy of deserving the qualification of science, but of art, where the best artist is the one who knows more and better security measures. I note that, once again, the saying is fulfilled that says: the truth offends. The book prologue says "Thus spoke Zarathustra" (paraphrasing the German philosopher Federico Nietzsche): When Zarathustra had said these words he contemplated the town again. He said to his heart: they make fun of me because they don't understand me. It is clear that I am not mouth to these ears. Will we have to break their ears so they learn to hear with reason? I see they have something they are proud of. That which fills them with pride they call "Manual of Security Measures", is what they believe distinguishes them from the ignorant.

Amunger

A word to the wise ...... give up while you're behind.

However clever you think you are, you are merely spouting incomprehensible gibberish to the rest of us mere mortals.   We all understand each other but we are not going to understand most of what you say because you use incredibly long words and put them together in a jumbled mess.  To misquote Morecambe and Wise - "you are using all the right words but not necessarily in the right order."

Edited by user 14 August 2019 06:49:54(UTC)  | Reason: Not specified

amungar - you have failed to grasp the key points and need to go back to first principles.  You are focusing on a formula rather than looking at the process...

If I can be bullish for a moment - 'routine' safety (high frequency low risk) in the main uses qualitative methods to determine risk.  This is based on experience so highly subjective.  

'process' safety and Occupational Hygiene (as an example) (low frequency high risk) in the main uses qualitative methods to determine risk.  This is based on tried and tested formulas, peer reviewed scientific research and industry validated information....

If I can also quote Winston Churchill - 'Courage is what it takes to stand up and speak; courage is also what it takes to sit down and listen'...

I am happy to continue the debate with you on PM if you wish...

Ah! Dave the answer to life, earth, the univerese and everything :0 

Or Sheldon Cooper?

Originally Posted by: amungar

In the same way that there is only one correct explanation for what has happened to a worker who has suffered an accident, regardless of whether or not one day we get to know it, we hope that there is also a single true version for the basic explanation of the operation of a labor system

I decided to brave reading what you are trying to get at. I think this sentence sums up why I would argue that you are wrong. You seem to think that there is an, as yet unknown, scientific formula that we could use to determine objectively what that actual risk is. If we only had this formula then we could all stop the guess work and health and safety would be a black and white- do this get, that - procedure.

However you statement above is completely wrong. It is the equivilant of asking a person what the most important part of a car is. Is it the battery, without which the car would start? Or the engine, without which it would run? Or even the wheels, without which it wouldn't move? They is no one objectively correct answer because it all depends on subjective views.

When a person has an accident there is not a single, correct, explanation. It depends on many interlocking parts that creates the sum. This is why root cause analysis should never have a single solution.

Look at it like the titanic. People died becasue it hit an iceberg, right? Wrong! they died because of the lack of life boats. Or because the rudder was too small, meaning the iceberg couldn't be avoided. Or because the ship was in water that was known for iceberg hazards. Or because the ship was going too fast. Or because the look out warning wasn't swift enough. Or many other reasons. Any of which would have stopped the mass death. There is not one correct reason for the accident, in the same way the is not one solution.

This is why we have the risk assessments. So we can spot the myriad of issues. Don't get caught up on a holy grail that doesn't exist.

Originally Posted by: O'Donnell54548

Ah! Dave the answer to life, earth, the univerese and everything :0 

Indeed! A true comedy genius and fellow humanist.

CptBeaky

You are one of the people who has supported the teasing of O'Donnell54548 and for this reason I will answer you only by saying something very simple (without derision or making fun of you): do not take me for ignorant by stating that I believe that there are only a few causes that explain the accidents. What you do not know and I know (and I have not said so far in this forum) is that there is a finite set of generic active and passive causes, perfectly determined, that are necessary and sufficient (that is, unique) to occur any labor damage (accident or pathology), if we exclude the particular causes that only exist in each specific case, which I call casuistic or “causal noise”, because no scientific theory can contemplate all the particular cases of any phenomenon.