Posted By Simon H I work with a very simple risk assessment model for some of our works and a client has queried something I have always done with it, I'd love some opinions. I have a 'Likelihood Rating' and a 'Severity Rating', multiply together to get a risk score etc... each are on a scale of 0 to 5 so the highest Risk acheviable is 25. Once the workplace precautions have been put in place this will (hopefully) reduce the likelihood of the event occuring but in some circumstances I have also reduced the Severity, and this is the issue that has been raised by the client, "how can you reduce the severity?" I use the example of a car that is 50 years old and a new car, better brakes better streering etc so you are less likely to have an accident (likelihood reduced) and it is also fitted with airbags, crumple zones etc so if an accident did occur the severity of the accident would be reduced and so would the Severity of your injuries; however he says that regardless of the car if you have a crash you could always die so the Severity should not cahnge! I see his point but have always felt that it is right in some circumstances for the severity to be reduced. Any thoughts would be well apprieciated.

Posted By Howard What about wearing a seatbelt. In the event of an accident the outcome would probably be less severe.

Posted By Paul Hay From a very simple point of view I have always said that generally control measures such as substitution, guarding, training etc. etc. reduce the likelihood where as PPE reduces the severity. Thus training someone to drive a car will reduce the likelihood of a car crash where as wearing a seat belt will reduce the severity of the injury suffered as a consequence. I suspect the person you are dealing with is applying the risk aversion theory and going a little over the top in suggesting the outcome of a car crash will always be a fatality. Regards, P.

Posted By Richard Altoft for any hazard you can have several levels of risk, so an untidy site might bring a near certainty of a trip and graze to the hand but whilst a death is conceivable from a trip at ground level if you fell badly and had a thin skull it is very unlikely. hence many combinations of liklihood at 0 to 5 and severity at 0 to 5 might exist In many of the cases the score will be the same (1 X 5 and 5 X 1 for example)But this exercise also exposes which are most worrying, which are most likely and which you can do something about. It also shows what action is worth doing and what benefits it brings which allows for AFARP approach and defence. But for your client to say "but death is always possible" is very naiave on their part R

Posted By Simon H Thank you to everyone who responded, its certainly clarified my thoughts on the way I approach my Risk Assessments. It was my first time using *this* websites forum, I've got to say it the best one I've been to by miles.

Posted By Adrian Watson Dear all, These questions are rarely simple. The are a number of difficulties around the concepts of hazard and risk. The first is that a hazard is not merely the potential of something to cause harm; it is the interaction between something, a person and the specific environment that produces the potentail to cause harm. The idea that a hazard is the potential of something to cause harm is an abstraction to the point of absurdity, as everything has the potential to cause harm. This means that the standard definition is meaningless. It is for this reason that a better defintion of hazard is "an event or condition that has the potential to cause harm". Furthermore, risk is the likelihood or probability that harm will occur; the severity or consequences of an event are actually independent of the likelihood or probability of an event occurring. This makes it impossible to multiply probability and severity together, thereby making the practice of multiplying the two together a bit of a nonsence. What should happen is that the probability of an event should be plotted against the severity of the event. From this plot, the most likely, and the most serious consequences can be found to provide boundaries to allow the assessor to determione what should deal with first and what is the most serious consequence that needs to be dealt with. Regards Adrian Watson

Posted By Adrian Watson Dear all, These questions are rarely simple. There are a number of difficulties around the concepts of hazard and risk. The first is that a hazard is not merely the potential of something to cause harm; it is the interaction between something, the specific environment and the person that produces the potential to cause harm. The idea that a hazard is something with the potential to cause harm is an abstraction to the point of absurdity, as everything has the potential to cause harm. This means that the standard definition is meaningless. It is for this reason that a better definition of hazard is "an event or condition that has the potential to cause harm". Secondly, the severity or consequences of an event are actually independent of the likelihood or probability of an event occurring; this makes it impossible to multiply probability and severity together. This makes the practice of multiplying the two together a bit of nonsense. A better definition of risk is the likelihood or probability that a specified level of harm will occur. From this the probability of an event can be plotted against the specified harm (severity) resulting from the event. From this plot, the most likely, and the most serious consequences can be found to provide boundaries to allow the assessor to determine what should deal with first and what is the most serious consequence that needs to be dealt with. Regards Adrian Watson

Posted By Adrian Watson Dear all, These questions are rarely simple. There are a number of difficulties around the concepts of hazard and risk. The first is that a hazard is not merely the potential of something to cause harm; it is the interaction between something, the specific environment and the person that produces the potential to cause harm. Therefore, the idea that a hazard is something with the potential to cause harm is an abstraction to the point of absurdity, as everything has the potential to cause harm. This means that the standard definition is meaningless. It is for this reason that a better definition of hazard is "an event or condition that has the potential to cause harm". Secondly, the severity or consequences of an event are actually independent of the likelihood or probability of an event occurring; this makes it impossible to multiply probability and severity together. This makes the practice of multiplying the two together a bit of nonsense. A better definition of risk is the likelihood or probability that a specified level of harm will occur. From this the probability of an event can be plotted against the specified harm (severity) resulting from the event. From this plot, the most likely, and the most serious consequences can be found to provide boundaries to allow the assessor to determine what should deal with first and what is the most serious consequence that needs to be dealt with. Regards Adrian Watson

Posted By Tom Doyle This is a very common problem with many risk assessment models. Two stages of risk evaluation are needed in order to get an accurate result from a risk assessment. The assessor must first determine a risk level by evaluating the hazard that is creating the risk as if there are no mitigation measures in place. During this stage a risk score can be determined by using severity, frequency of exposure, probability of avoiding or limiting harm, and the probability of the occurrence of a hazardous event. In a properly constructed risk model, the product of these four factors will provide a relatively accurate prediction of the likelihood of the occurrence of harm. Once the first stage has been completed risk mitigation measures can be utilized in order to reduce the probability of harm. Whenever possible, risk should be mitigated by reducing the probability of the hazardous event and/or the severity of harm caused by the event. If the hazard cannot be eliminated, or its severity greatly reduced, engineered controls should be employed in order to reduce the probability of exposure to harm caused as a result of the occurrence of the event. These types of risk controls will normally not affect the severity of harm caused by the hazardous event but will often greatly reduce the probability of injury due the occurrence of the event. Even with well designed, and reliable, engineered controls in place there is usually some degree of residual risk that must be mitigated though the use of instruction, training, PPE, etc. For example: Modern day aircraft are designed with very reliable controls that will help to ensure that the aircraft will stay in the air throughout the intended duration of a flight. Thus the plane is not likely to crash (hazardous event) due to a malfunction of the equipment. That being said, we only allow trained and competent pilots to operate the equipment as a further risk mitigation measure (training and instruction). With all of these mitigation measures in place we still require flight passengers to wear seatbelts during portions of the flight (PPE). Passengers also receive preflight instruction informing them of the procedures to be followed in order to avoid or limit harm should the hazardous event occur (training and instruction). Still, even with all of these mitigation measures in place, planes occasionally crash. People who fly on planes should be fully aware that this possibility always exists. It would seem that they are willing to take the risk associated with air travel in order to reap some type of benefit as a result. It would also seem that, in general, the public is willing to tolerate the slight probability of harm in order to reap the benefit of reduced travel time even though the consequences of a hazardous event remain quite severe (broadly tolerable residual risk). Cheers Tom Doyle Industrial Safety Integration

Posted By Dave Wilson Agree Adrian but a bit long winded mate! Remember its the 'likelyhood' of something happening not worst case scenario! If the likelyhood is death then you shouldnt be doing it!!!!! ie fall on flat surface bang head = death - so remotely removed! more likely it would be the Japanees sniper scenario - get up, brush of and pretend that no one saw you! unlikely

Posted By Andy Walker Fully agree Vince. For the front line troops the simplest and easiest method works best. As was mentioned earlier in the thread its the thinking about and awareness that is more important than the method. Andrew

Posted By Vincent Shields I believe, generally speaking, that you are correct. Doesn't mean it won't happen, but rather than it is 'very unlikely'. Any risk treatment would need to be proportionate to that negligible risk. Yes, there are lots of other variables, but we're talking principles, not details, in this thread. My view anyhow, Vince