Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
2010 Q5 Configuration Control
#1
An attempt for comments please
- not timed and probably too short. What might be missing?


Attached Files
.pdf   IRSE-Mod1-2010-Q5-DAP.pdf (Size: 2.15 MB / Downloads: 103)
Reply
#2
(08-09-2016, 04:40 PM)dorothy.pipet Wrote: An attempt for comments please
- not timed and probably too short. What might be missing?

Definitely too short; the harder thing is to know how to expand it.
Do remember though that it was set as a 20 minute question.

It is not one I would have chosen as it seems too limiting; as I summarise it:

for FOUR phases, describe ONE hazard with ONE consequence and ONE means of prevention (different in each case).

In truth your answer already seems to have multiple items within most boxes which could be argued to be non-compliant with the wording.

As I have mentioned in another of the 2010 questions, I do feel that what you list as "HAZARDS" are actually CAUSES.  In order to distinguish the two, I find it helpful to define a "system boundary" and then see if the item could be a condition or state at that boundary which would, in a particular combinations of circumstances then lead to some accident.  So of itself, water is not a hazard; conversely it is an essential component of life.  However a shallow pool of water on a road that could then freeze would certainly be a hazard to an incautious person walking or driving over it unawares, similarly if water starts dripping into equipment usung high electrical voltages then it is clearly a somewhat different hazard. I think that usung the words IF or WHEN in the description is useful to explain when the item listed becomes hazardous.
  
Perhaps therefore the way this should be answered is to choose a true hazard in each case and then go on to describe at some length, including a range of causes that could lead to the same hazard and consequence (given that a hazard is a condition that could potentially lead to an accident, then the consequence would obviously be the (or one of the) associated accident(s).  The problem with this idea is that you clearly need to find different hazards and consequences for each lifecycle phase and I don't see how you can.

A hazard is "a condition that can cause harm" so perhaps we could first consider the different type of harm (which would be the consequences asked for) -
a) injury to passenger as a result of a train collision,
b) ill health to employee from exposure when undertaking their work,
c) disrupted train service,
d) financial loss,
e) contaminated environment.

We'd then need to select a hazard that could lead to that consequence.  I might be able to get a), c) and d) to flow from a lack of version control, but I think I'd struggle with the others, so that doesn't seem to help much.
In fact a good thing about your answer was that you did have a consequence in each of these categories of harm; however I think you should have made this more explicit and obvious, given that this was a module 1 answer.

Also it is really the individual CAUSES that could potentially be prevented; the HAZARD is only eliminated if all the possible causes are prevented.  Obviously a HAZARD can be present but MITIGATED by placing certain barriers to prevent it resulting in an accident, but that is not what the question asked.

So I am at a loss to know what to suggest.  I can't see how to answer it precisely within the constraints given; indeed I don't actually think that it is answerable and to me I think it is a poor question.  I may be right, I may be wrong but for me in the exam I would decide to AVOID (if I am correct it is unanswerable; if I am wrong then clearly  cannot answer it as the examiner wants).

If we work on the assumption that it was a sloppily worded question and what the examiners really wanted was a range of causes of an undesirable situation and to describe how four distinctly different means by which some could be prevented (which is effectively what you did), than I think the question is far more do-able.
Perhaps the descriptions needed to be longer and explain more fully how an error in version control could lead to the outcome; in which case the decision to present in tabular form probably incompatible with such an approach.

Freed of the space constraint then could have given far more detail in each case to make it a concrete example, rather than being somewhat abstract.
A certain resignalling scheme on the south coast would have been a wonderful exemplar.

1. Scheme Plan version A showed Green /Red from existing signalling to "final" scheme; this used to mark out axle counter head positions by the installation.
2. Scheme Plan reissue version B added 40 new axle counter sections to correct initial errors.  This showed Green /Red from existing signalling to "final" signalling, but there was no definitive list of changes from version A.
3. Installation realised they needed to add the extra heads and re-label all the ones already marked (and sometimes installed). but they didn't recognise that just a few of the original ones in all this change had actually moved by some crucial metres.
4. Testing function had placed version A on the Scheme Plan on the cabin wall and although they had updated version, started marking up progress on that one,  Since there was no major change and it was good enough to record which signals and points had been tested by the location team, its use for this was perpetuated.  The separate axle counter testing team had not been issued with a suitable plan, so they found it a handy document to refer to.....
5. There were multiple versions of draft Axle counter siting forms and the formal ones were issued very late.  When cross-checking was done to these, the error was not spotted since the form referenced the head to a new signal position, yet the intended position of that signal had itself moved during the evolution of the project and the axle counter form reflected where it was to have been positioned, not where it actually had been placed.

The net effect of this was that both limits of a certain section were in the wrong position by about 10 metres.  Just before commissioning, it was fortuitous that someone noticed that a head which should have been proving clearance of a crossover was actually foul ad therefore a collision could have resulted.  This was test logged and corrected.  However no-one went back and rechecked that the track section was still over the defined minimum length of 18.3m; the test certificate had already been signed off as ok. 

Hence when commissioned, the section had the potential to be showing clear when occupied, as it was straddled by a coach with one of its bogies counted out of the far end of the section before the other bogie had been counted in.
Therefore the correction of one end to avoid one WSF was undertaken without thinking of the effect on the other end which therefore caused a different WSF.

This true little (but abbreviated) story illustrates that the incident arose partly because of lack of proper version control in the first three phases.  It could almost be used to illustrate the last as well; except that axle counter siting forms were at that time not even regarded as records which should be given to the maintainer and so they had access to nothing definitive to tell them where to put back any axle counter heads which have to be removed for P'Way activities etc......

I think that I could readily suggest means to prevent the various causes and therefore attempted to answer the question in such a manner, but I do wonder if that is what the examiner really envisaged and therefore how they would allocate marks

===================================================================

Going back to your answer: I think that having done what you did, I'd perhaps have decided I needed to extend and the best way to do that would probably be to put some numbered references on an entry in the grid and the on A4 paper put a bit of "flesh on the bones" of the skeleton.  That may have made the best of it; however I am still of the opinion that the wisest course of action would be to have selected a different question.
PJW
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)