Re: NRC <> Brisbane to Sydney safeworking.

["David Proctor" <daproc.spambait@umpires.com>]

Richard Wardle wrote in message <5lB32.100$ih1.327@nsw.nnrp.telstra.net>...

>The system is suppost to be fail safe. If the computer stuffs up then the
>signals are set to STOP.

This is what is supposed to happen - but again, we are talking about
machinery stuffing up here, and what happens if the bit that makes it
failsafe is the bit that actually stuffs up?

>occur.  Safety is the NO 1 prioity for an engineman.  It is instilled in
him
>from a trainee so the safety of his train is paramount.


Not disputing this - I know it for a fact from my dealings with friends who
are drivers.

>That is rare on the railways as the driver knows where the signals are
>located and he knows how far they are away.  He can judge the speed of his
>train and the time it will take to stop his train.  Misinterpretation is
>usually due to other factors like fatigue, obstruction or lack of
>maintenance of the signal itself eg. dirty lenses.


The same applies for train orders - an order to proceed to Xxxxx loops would
only be misinterpreted due to fatigue, etc. The other factors, obstruction
of the signal, etc are no longer applicable.


>If there is a communications breakdown at the moment the CTC system can
>still carry on if the signalling system is not effected.  If the signalling
>system is effected then they can bring the signals under local control so
>the system can still operate by manning the signal boxes (like they have
>done for the last 140 odd years).


Depends where the communication outage is - if the system still works
properly, but the link with the control centre is down, then yes. But that
is not the problem the whole time - hence the reason why they still retain
the half pilot staff's at each end of the section.


>I disagree there are a lot of safety issues to be addressed.  So it works
>elsewhere.  What changes to their systems did they do to bring train orders
>in.  My understanding is that the SRA wishes to keep the current
>infrastructure and still bring in the working.  Computers don't make
errors.
>Programmers do.  If there is a bug in the computer program then it can have
>fatal concequences.

This argument can be applied to computer based interlockings on conventional
signalling systems too - yet there is no claim from drivers regarding
reduced safety.

David "The Doctor" Proctor
daproc.spambait@umpires.com

>
>
>>
>>David "The Doctor" Proctor
>>daproc.spambait@umpires.com
>
>Richard Wardle
>
>http://www.geocities.com/TelevisionCity/Network/1104/index.html
>>
>>
>
>