2013 Q3- Authorising past signals at danger

[PJW]

Another I have been sent recently; not sure if I am going to find time to review in remaining time so posting here in case anyone else gets the chance

[PJW]

unfortunate /

Another I have been sent recently; not sure if I am going to find time to review in remaining time so posting here in case anyone else gets the chance

Some comments.

1. As with most of your other answers, I liked the clarity of presentation, including the structure of your nested lists like folders and sub-folders.  Both clear and tome efficient.  However in some cases may be worth adding numbers after your arrows in a question where a later part needs to cross-reference back.
Also inclusion of little sketch very useful and the table later on was well planned. So much easier to follow than most; creates a good impression even before start to read.

2. You start by listing reasons why signal might be at red (all seem to be failure conditions except for the last one) and therefore signaller needs to talk past.  Not explicitly asked, but could certainly contribute to a good answer.  However to make the best use, then linking the items in the "SPAD occurs" / "Potential Risks" lists back to that first list where appropriate would tie together.
Also I think you should have presented items 1-5 as sub-bullets all examples of a signaling failure, but you should have added a range of more different reasons; for example:
a) animals / trespassers reported on the line
b) following a "bridge strike" by a road vehicle,
c) having in abnormal working to make a permissive move into a platform either where one does not normally exist or where signalling controls are designed that signal should not clear,
d) axle counter section recently reset so Aspect Restriction is in place
e) an engineering train to proceed up to PICOP's protection (detonators and light) prior to entering possession.
 
Another good list of things that are likely to increase the rate of SPAD at a signal.
One of the things that you missed that is very relevant to this question was anything related to the signal in rear; for example this could be uneven signal spacing so that after a series of widely spaced signals, then the driver encounters a Yellow that is only barely braking distance from the Red.  
Another example is that the signal is SPADed not because its sighting distance is poor but because it was encountered unexpectedly as the one in rear had poor sighting and the driver wasn't looking at the critical brief period it came into view and was unaware that he'd missed it.
Or the case when he had seen it ok, but didn't really worry that it was Yellow because needed to slow down anyway for a station stop or Temporary Speed Restriction observance- the problem being that once that reason to stop / slow has passed then driver able to accelerate again UNTIL suddenly encounters Red with what might seem like zero warning.  Given the wording of the question, then suggest these reasons would have been good ones to include.

Obviously for a SPAD to occur it absolutely needs
a) a train moving towards the signal and
b) that signal to be at danger,
So a busy line and a signal that often has to be at red to protect a conflict are also factors that might cause a poor SPAD history, whatever the "SPADability" of the signal itself and it is important to keep that in mind as well.

3. HOWEVER you were asked to discuss RISKs associated with such authorization to pass at danger, and it does seem too much background before getting started on really answering the key issue.
You were not necessarily completely wrong; DISCUSS RISKs must mean:
a) talk about what hazards there are following that authorisation (i.e. what could possibly go wrong)
b) give indicative quantification of the likelihood of that thing going wrong and the factors that would affect,
b) talk about the consequences that could result from that thing going wrong- including what sort of accident and what sort of losses might be suffered and how big are they likely to be.
However I think you focused too much, too early on the relevance of the signal with the SPAD history; I'd have worked that in later.

Given the fact that S3 at danger for whatever reason and ignoring for the present its SPAD history and assuming driver does stop at it, the basic things that could go wrong:
a. Signaller misunderstands where that train really is / selects the wrong signaller route card and follows those instructions perfectly but misguidedly
b. Signaller gets that right but overlooks a set of points within the route for example
c. Signaller when operating points on IPS accidentally sets one the wrong way and fails to realize
d. Signaller unable to move points from IPS and so has to rely on "point setting agent" to wind points by hand on site.  That person might wind the wrong points or might wind the correct points to the incorrect lie. Could be due to miscommunication of the instruction or perhaps person on site got confused between left and right.  Or perhaps to save valuable time another person comes to help to do some of the point winding but there is confusion between them and an end gets overlooked- a slip connection/  switch diamond / swing nose crossing are the classic types that get forgotten.
e. Perhaps all the above goes right, but when signaller contacts driver, he makes error and mistakenly rings one of the other drivers waiting at red signals in the area; therefore authorizes the wrong train to move compared with the way that points have been set.
f. If signaller correctly conveys info  to the right driver to pass the right signal when it is safe to do so, there could still be room for error.  Did the driver actually understand WHY being called past a Red; did the message that needs to be very careful and inspect a portion of line get properly assimilated and when driver gets there will he remember?
g. Does driver understand correctly any speed restrictions with which must comply and exactly how far have been authorized to proceed; perhaps they think they are being authorized all the way to the exit signal, when the signaller intended that they were just to go a short way "to get behind" another signal at which they would reverse direction to be able to go to another line.  A classic is a driver looking to stop at a specific numbered GPL that they expect that they will see in front of them so carry on until they find it, when in fact the  GPL actually is only after a 100m and it  faces the other way being the signal that the signaller intended next to set for them once they had reversed.  Or perhaps there is a failure of a number of signals and the driver told that he should pass 8 at danger, but miscounts and SPADs the one that was still to be observed.

4. Looking at your list of Potential Risks, in most cases you have listed ACCIDENTS and in some cases have hinted at LOSSES.  Amongst them though is number 5 which is a HAZARD (i.e. something that could cause an Accident); ditto number 9 at bottom of page and number 10 on next page.  Then you list some other LOSSES.
Lots of good content, but did need a bit more clarity.

I recommend that a column table approach as in the next section would have been the best way to organize here.
Columns for: Cause / Hazard / Accident/ Losses
Columns also for: Likelihood (L/M/H) and Consequence (L/M/H).

Basically Accidents are:

1. Head-on collision
   
2. Sideswipe collision
   
3. Rear-end collision
   
4. Derailment at point or from overspeed
   
5. Collision with road user (pedestrian or vehicle) at level crossing
   
6. Collision with trackside worker
   

Similarly Losses are:

1. Death
   
2. Injury / long term health issue
   
3. Damage to equipment
   
4. Damage to environment
   
5. Loss of service
   
6. Direct financial loss (costs of remedial / consequential working, compensation, loss of revenue)
   
7. Loss of reputation (political and longer term financial impact)
   

There is going to be quite a lot of repetition so if you number the items and provide such list, then you can use such numbers to populate your table with the columns; each row would look at a particular hazard (that I have basically put in item 3a-g not have not worded n the form:
if <undesireable condition> exists, then if < careless action / unfortunate circumstances > then <accident> might occur
but hopefully you can see that this does DISCUSS THE RISKS.

Perhaps then is the time to introduce the SPAD record.
Let us assume that we are talking about signal S3 at Red.
If S3 at red, then there is a chance it will be SPADed; if it has a poor SPAD record then this means that we can assume that the chance of that SPD occurring is higher than average signal.  So now we think what effect that this has and this will depend on WHY that signal was at Red in the first place.  If it is protecting track workers then any SPAD at it would endanger those workers- HOWEVER this does not fit the question since signaller would not be about to authorize a movement past it.   Conversely if it is protecting a possession, then that would fit, but how much risk there is would depend on the sort of SPAD.  Likely to be a significant distance (400m perhaps) between protecting signal and PICOP's railway, so if the SPAD is due to slight mismanagement of braking (many are) then overrun may only be of a few tens of metres and hence quite low risk.  Conversely if the SPAD is due to failure to take cognizance of signal in rear, then the SPAD could be for a very considerable distance and get into the possession when it was unsafe for it to be there (other trains, missing track etc.)

If we now consider S1 has the poor SPAD record, then in the circumstances when a signaller is procedurally authorising past S3 then it is a rather different scenario than normal.  Train speed should be quite moderate and driver particularly on their guard, so the chances of SPAD'ing S1 must surely be lower. Any issue of uneven spacing of signals is now irrelevant; similarly can hardly have "missed the distant".  If the power to the location has failed so S1 is blank and the TPWS dead, then it is certainly possible that driver could miss it entirely, and so perhaps in that particular scenario the risk is higher than normal (although the AWS horn really should have alerted the driver).

That is the sort of thing that I think you should have included in your discussion and involved the i) and ii) when there is a salient distance between them.   Whereas you had two sets of good ideas that did have their relevance tangentially, you rather answered as if that were the main question.  You had the various elements and you would have got marks for that; what you didn't do was join them in such a way really to answer the question part a).

Part b)
Only a quick look, but this seemed far better answered.