2008 Q7 RISK ASSESSMENT OF SINGLE LINE WORKING

[mark bedford]

I've attempted this question

Please give some feedback
Mark

[PJW]

I've attempted this question

Please give some feedback
Mark

The question is worded re "modifying the railway"; hence I assume that there is must be some physical change. Indeed I'd interpret as some form of SIMBIDS or reduced capcity wrong direction signalling; I have difficulty with your interpretation that suggests purely procedural wrong line working. I suppose you are saying that the modification is the installation of the crossovers and so I'll go with it but perhaps am a little dubious and not really what I was expecting so your answer must win me over.

I don't think there is value in quoting so many names ans numbers; I'd be briefer and just state an NR environment; certainly don't tell me the standards that are not applicable (and always better to give their nomenclature than a reference number). Also I think you'll find that the Rule Book is a Railway Group Standard and so "owned" by all the members of that group, RSSB merely being the agency that is the "custodian".

I do like the tabular presentation which is both clear and demonstrates that you are familiar with the way risk assessments are often presented.
You have chosen to go further than was asked in categoriing severity, liklihood and risk but that is no bad thing and again demonstrates knowledge and experience. I don't think though that I'd have put numbers in; High / Medium / low would have been as far as I went. You also haven't explained what those numbers are and have left the examiner to assume / guess. Further it appears that you have ADDED severity to likelihood to get risk; to me you'd be losing some of the credit that you may otherwise have gained. You'd have been better rating the risk as Intolerable / Tolerable / Broadly Acceptable.

However rather than just launch into it, a short paragraph introducing it, defining terms and explaining how it addressed the question would have been more valuable than most of your initial one-liners.

You probably ought to have explained a bit more about your two sets of columns; it didn't help that the figures with the additional controls are sometimes not reduced from those with the existing control.

Look down your first column headed "hazard". A hazard is strictly "any situation that could contribute to an accident" and some of the entries seem to be the accident itself. Others are factors that may prevent the effectiveness of what could otherwise be a partial mitigation after the accident has happened.

It does seem very odd that a railway would bear the costs of these crossovers and yet really implement "only give authrity for use as last resort"- it surely must intend to use it on a reasoable number of occasions. If it really was to be quite rare then surely the existing crossover at A and just one at B would suffice- yes there is a further time penalty and additional riskl by needing to reverse a train between runnuing lines, but if it is hardly ever going to be used anyway.......

You have certainly considered both staff safety and operational safety as you were requested to do; so good.

I don't feel that your format gave you room for enough emphasis on the "describe any systems which may be provided" part of the question.

Certainly more here than you could do in 30 minutes.

Overall impression is certainly someone familiar with railways and the sort of things that are normally put in as risk assessments in various NR documents to keep the "safety parking wardens" happy. Not so convinced that it was a good answer to the actual question; it'd be a Pass I think as so much better than so much else that the examiner would see but you wouldn't do as well as I think you might have expected.

It'd be interesting to see if others concur with my view or think differently

[mark bedford]

I've attempted this question

Please give some feedback
Mark

Thank you Peter for your very helpful feedback.

Interesting how a question can be interpreted in different ways. I'm now thinking I should have made and stated some additional assumptions at the start of the answer. It would seem logical as you say to provide some sort of signalling system as this is a signalling exam. As this module covers ESM then stating what stage of the system life cycle this assessment covers is probably also important. The question says this is a proposal, so this is likely to be the concept & feasability or requirements stage.

Defining Hazard as a contribution towards the accident (rather than harm) leaves me pondering. The environment of the tunnel has to be a major consideration in the event of an accident and a consideration in the answer. Should further hazards after an accident has happened, i.e. fire, emergency access and egress, be in a separate risk assessment to the operations and staff safety? I hadn't considered that before but it opens another mind field of organising different hazard analysis to get to the final conclusions of the feasability study. I've no idea how this works in real life or what will be in the examiners mind! So I guess state some risks if a accident does occur but be aware not to over state these risks.

If this is a feasability then the conclusion of the analysis could be that this proposal is 'not feasable'. Therefore the assumption to only provide emergency crossover with no signalling may not have matter if I had then proven through the hazard assessment that this is a poor proposal. Probably best for the exam to use a system that would be acceptable.

Thamks for you help

[Peter]

Defining Hazard as a contribution towards the accident (rather than harm) leaves me pondering. The environment of the tunnel has to be a major consideration in the event of an accident and a consideration in the answer. Should further hazards after an accident has happened, i.e. fire, emergency access and egress, be in a separate risk assessment to the operations and staff safety? I hadn't considered that before but it opens another mind field of organising different hazard analysis to get to the final conclusions of the feasability study. I've no idea how this works in real life or what will be in the examiners mind! So I guess state some risks if a accident does occur but be aware not to over state these risks.

The main comment that I was going to make about your list was the fact that, as you have come to realise above, many of them were secondary events, assuming that some initial event had happened and hence may in fact affect the "severity" of the first event but not its likelihood. Hence in reality, more detailed analysis of which events are dependent on others needs to be assessed to work out the overall level of risk (way outside the scope of this question).

Peter

[bigcheese]

Mark
Hoping to make a helpful contribution rather to leave you feeling that I am jsut being picky I have lksited your hazrds below with comemtns that may illumninte the points that PJW has made.

Most importnatly it is absolutley key to define "the system" as the hazards will lie on the boundary of that system - whatever it is - I have taken your definition of work as the basis.

Head on/ rear collision during single line working - this is an accident i.e one example of the possible harm that can arise from the hazards inherent in the system / process we are studying. It is probably the most harmful outcome but not I woudl suggest the most likely. We use the concept of risk to help us "measure" its importance.

Fire in Tunnel after collision with derailed train - another outcome (though fire is an awkward customer that doesnt fit the EN / YB model very well). Certainly it is outsdie the control of your process so it is not a hazard - it may be best to deal with it in the casualty modelling.

Difficulty of egress in an emergency - this is going to effect the severity of the outcome (the amount of harm) and probably cannot be influenced by your process so it is not a hazard.

Lack of access for emergency services - as the previous one, though you shoudl be taking this into account b the mitigation measures you design into your process/system. The harder it is the more reasonable it will be to spend some moeny and effort in makign it easier.

Emergency exit route blocked in both directions by train wreckage - this again influences the outcoem, in this case for worse.

Confusion who is in-charge of single line working - this now soudns liek a cause (or causal factor) as it is within your control to get this right. if it is wriong then you amy be lucky and get away with it otherwise there will be some degree of escaltation.

Rest assured that this is not easy adn EVERYONE gets it wrong and there are often endles debates about waht is a cause/hazard/consequence. however it is achievign this clarity early on which is at the root of designing a safe enough system or process. This avoids the courtroom debate about I thought you meant...or I didnt realsie that was included...etc. This kind of thigns is called fault and event free modellign in YB - the faults or causes (because you can have hazards even when everything is working proprerly) lead to hazards the consequences of whcih can be modelled in an event tree. This is not the only way to model - outside the UK people and in other industires peopel use Bowtie diagrams with a rather more visual and very good at modeling the barriers pr mitigations.

As to what really are the hazards in your process / project I would suggest things like:
- erroneous but valid movement authority given to train
- deliberate disreagrd of movement authority
- distraction (may be a cause....)
- train fails or separates in section without notification

These are rather general and woudl need a bit more flesh on them for the exam. Your consequnece column looks OK but remember there is always the nothing happened today outcome (usually the most frequent). Hope this helps