h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

I work for a uk steel producing company who has a policy whereby every job has to be risk assessed.

This was originally done on a ticksheet where you had to mark off the possible risk the job would have under 50 or 60 odd headings such as 'danger to contractors', 'safe access', 'working at heights' etc etc. The risks were categorised from 'danger of death' down to 'improbable risk of injury' and were scored accordingly with the final score giving the overall risk for the job.

The ambiguity of the whole system made it laughable. For example, if I had to put down a risk for working at heights I could think to myself 'well the worst that could happen is that the scaffold or safety harness will fail and I will fall to my death' so the job has a (all be it small) risk of being fatal - after all, working at height on a scaffold platform is not without ANY risk. But NO, the powers that be say that that I am at little risk because I am on a scaffold and I am wearing a harness. But I am at a massive risk compared to doing the same job at ground level. Its all relative!!!

Which brings me onto my point about risk and playgrounds, which someone pointed out to me whilst getting bogged down filling out one such assessment....
Every piece of moving machinery at work has to be guarded off and if you want to remove the guards, you have to immobilise the machinery and isolate it electrically or mechanically so there is no risk of being trapped/crushed by it.
Lets just for minute apply these rules to a childs playground and try and imagine what IT would look like if all the moving parts were guarded off and risks of falls were barriered off.......

I think that awareness, alertness and good old fashioned common sense beat trying to analyse the unquantifiable. Just my thoughts anyway.........

I can appreciate your position. One of the chief difficulties of risk assessment is its misuse, possibly due to its misunderstanding, by the decision-makers (read: management) for whom we all work. Another is the fact that risk assessment is still rooted firmly in the unknown - in fact, in the unknowABLE if we have our way (in other words, we would rather everything we say COULD happen would actually NEVER happen). It's a strange and unnatural way to think, and that leads to a lot of problems when interpreting risk, and in fact just about anything having to do with probability.

In the case you describe about working at a height, you are correct in saying that there is a risk of falling due simply to working from a height. This risk is natural, and should be accepted - it's the only way to get the work done, after all, unless we want to use robots for everything (which would, of course, introduce a whole new set of problems in both risk assessment, cost-benefit anlysis, and even ethical concerns - why get rid of a perfectly good human worker anyway?).

You're also right about the evils of the infamous "checklist" method - probably the most commonly used method across industry itself, despite its many limitations. But above even that, risk itself is a muddled thing, as you've noted, and prone to a significant problem: managers frequently pay too much attention to the "most severe" consequence events, at the expense of paying attention to the "less severe" but more likely events. That's why risk assessment has BOTH consequence and frequency, and I think that gets missed way too often in the most common risk analysis misapplications in industry. Just as you pointed out.

In my work, the involvement of the guys that actually DO the work is imperitive and required - partially, at least, for us to be certain to capture and prevent such illogical loopholes that can lead to problems like the one I described above before they are implemented. Another good point you bring up is one that I've been harping on for years - risk control is too often simply not applied logically and without bias. Usually, the reasons are ultimately political - accidents involving workers in a nuclear laboratory, for example, are so undesirable due to political attention and governmental scrutiny, even those involving a sprained ankle will have extraordinary measures taken at perhaps an equally extraordinary expense to minimize such risks - hazards that are arguably due to that fact that one happens to be a living person that walks to the bathroom to blow one's nose, and have nothing to do with working at a nuclear facility at all. This frequently has to do with how statistics regaring injury and accidents and incidents at a nuclear facility are tracked and presented by the media and/or other parties that wish to shut such facilities down. Hence, the problems of politics intrude upon the actual safety of the workers themselves. True, and probably inevitable in our society.

While that example may seem extreme, it is in fact a real one. Your example that we pay far more attention to people not being injured on the job than we do with out children's playgrounds is not a particularly fair comparison (the probability of losing a finger to a lathe, for example, is far more likely than from a merry-go-round, I would argue), but the point you're making is valid nonetheless. Do I blame politics for this state of affairs? Well ... yes and no. Politics may drive a lot of super-risk-aversion, and we may live in a very risk-averse society due to a variety of reasons, but that doesn't mean we have to make illogical decisions about controlling risk.

"I think that awareness, alertness and good old fashioned common sense beat trying to analyse the unquantifiable." Well, the real goal of risk assessment is its application and control - yeah, I agree awareness and alertness and common sense play the major role of humans keeping themselves safe, but one needs to know just how important those steps are in case someone else changes them unknowingly in the future. In the examples you give, the application of too much risk control (not to mention with a blunt tool that isn't really all risk assessment in your field) can be counterproductive. I have many collegues that feel the same way about "safety intruding into their work," but unfortunately I can name just about as many others that work in the regulation field that feel more control over the workers themselves is necessary to prevent these "common accidents that make us look like insane daredevils." I would also mention that whenever something does go wrong, it is usually the company/government that is the first one to be sued - right or wrong, this perceived vulnerability to the "injury lawsuit" drives a LOT of safety regulations in industry. What's bad is that such strictness can and does backfire, and winds up being worse for overall worker safety.

It's a complex beast, this risk assessment!


- JD

Fair comment.

I think that in the 'awareness and alertness' bit I should have included 'experience' and 'process knowlege'.

The last couple of points you made were true. I often wonder if the signed risk assesment is a way of 'pointing the finger' about who to blame after an accident has occured.

Maybe just me being cynical