New Every Morning

[whole diary]

[onthecushions]

I've started this topic as a place to record on-going goings-on between Reading and Paddington as there are many of these and they seem to start a new topic each time. My apologies if I have missed someone's pre-existing topic, and also to the hymn writer.

Tuesday and Wednesday, 6th and 7th October havn't been good FGW^▸ (First Great Western) days so far.

Tuesday evening saw a points failure East of Slough that had all down main (West-bound fast) trains stopped. My 1921 ex-Paddington reached Reading at c2055.

Wednesday had amongst other things an HST^▸ (High Speed Train) with a broken windscreen, limping into Reading with other services following slowly. The relief lines were then cleared for it to run direct to the depot at OOK. This caused a jam on the up-main (East-bound fast) as all stopping services were diverted onto it, well... stopping... All the passengers (customers) on the failed train were crammed onto other services from Reading for a 50 minute crawl. I saw the empty "failed" train zip past the queue of crammed standing services at about 60mph...

On both delayed trains the Guard (sorry Train Manager) used the PA^▸ (Public Address) promptly and in an exemplary manner. However the Tuesday service while in contact with "Network Control" had no real information and the TM^‡(resolve) was as suprised as his customers when the train moved off. Also the platform information at Reaing was nonsense as trains were shown as on time until 2 minutes or so after they are due, when a new, later departure was advertised. This continued in a most exasperating manner. As these indictors are supposed to be linked to signalling, does this mean that the signal-persons don't know where their trains are, either?

The old monitor indicators from BR^▸ (British Rail(ways)) days were much more accurate and helpful, as was the now defunct Solari indicator on the concourse. Please bring it back, and BR (InterCity And NSE^▸ (Network South East)) as well. They were better value.

At the same time as this confusion occurred at Reading, the announcer requested customers to nominate staff for good service awards. That at least had people smiling.

OTC

[Lee]

I've started this topic as a place to record on-going goings-on between Reading and Paddington as there are many of these and they seem to start a new topic each time. My apologies if I have missed someone's pre-existing topic, and also to the hymn writer.

Its a matter of personal choice, really. The reason I post such items as seperate topics is because I date them, and I find it easier to refer back to what happened when by doing it that way.

Whether members wish to post such items as seperate topics, or have a single "running topic", then either way is fine.

I'll pass your apologies on to the hymn writer when I see him, OTC 

[IndustryInsider]

Also the platform information at Reaing was nonsense as trains were shown as on time until 2 minutes or so after they are due, when a new, later departure was advertised. This continued in a most exasperating manner. As these indictors are supposed to be linked to signalling, does this mean that the signal-persons don't know where their trains are, either?

This usually happens when a train has stopped somewhere for a considerable amount of time. For example, if it was stopped just after Slough coming from London but left Slough on time then the CIS^▸ (Customer Information System) systems don't update it as delayed until it moves past the next reporting location. Depending where in the country it is this is either automatically linked to certain signals, trackside sensors, or in some cases still reported manually by signal-box staff.

When the train moves again and passes Maidenhead for example the system will assume it is going to run full whack from there to Reading and will show on Reading screens as due in ten minutes time. This takes no account as to whether the train is moving slowly in a queue and is more likely to take 30 minutes, hence the minute-by-minute increases with no train in sight.

Whilst the level of automacy is very good at dealing with on-time trains, or single delayed trains that are now proceeding normally, making announcements and displaying largely accurate information on screens. But as soon as a delay develops to the stage where several trains are involved or trains are stopped completely its limitations get shown up.

It doesn't help that this is the very time when staff and customers need the system most. In the old days a member of staff at each station with its own native CIS system (usually the station announcer) would be able to make manual entries and take into account the factors I have mentioned above. Now, with a mere handful of people attempting to control the whole of the FGW^▸ (First Great Western) area this is simply not possible.

[onthecushions]

Thankyou, II, for the explanation. I believe the Customer Information System (CIS^▸ (Customer Information System)) was originally Passenger Information System until the acronymn was noticed.

This morning the CIS, perhaps in embarrassment over these posts, was blanked out with a "problem".

As you say the CIS is mainly needed when there is a problem but as its displayed times assume that there isn't....this is FGW^▸ (First Great Western) after all.....it worsens the customer experience of delay.

Usually the PA^▸ (Public Address) announcer at Reading is fine. As Reading is one of the very big non-London stations with a footfall of over 14M/yr, (ORR^» (Office of Rail and Road formerly Office of Rail Regulation - about) figures), it surely does justify a human touch to the computerised CIS (if only to switch it off!).

Otherwise, this morning (for me) was wonderful; clear blue sky, best trains, route and staff in the UK^▸ (United Kingdom), all clean and on time.

OTC

[Boppy]

Thanks IndustryInsider for that info.

I commute each day between Reading and Paddington and have always wondered what/how the system based it's estimates on.  I'm pretty much in agreement with what has been said that the system really needs to show it's usefulness when there are large problems and not when things are running happily.

One small niggle that maybe someone can clear up about the system.  It's always perplexed me how the system will say the 09.35 is expected ontime at 09.35 but the clock running underneath says 09.38 - is there not a simple check to push the expected time on so that it is at least what the current time is as it's minimum value!  I am assuming the current time comes from some central point in the CIS^▸ (Customer Information System) system.

Cheers!

Boppy.

[Tim]

One small niggle that maybe someone can clear up about the system.  It's always perplexed me how the system will say the 09.35 is expected ontime at 09.35 but the clock running underneath says 09.38 - is there not a simple check to push the expected time on so that it is at least what the current time is as it's minimum value!  I am assuming the current time comes from some central point in the CIS^▸ (Customer Information System) system.

Technically that would be (or ought to be, surely?)a very easy thing to do.  It would give a better impression to customers and make the estimated times at least a few minutes more accurate. 

[IndustryInsider]

I'm not sure if this is still the case, but if memory serves me correct, when the Thames Trains system was installed about a decade ago, there was a 3-minute threshold built in to the system for some reason. This meant that a delay of one, or two minutes would never be shown. A train would either be on time or 3+ minutes late. Not sure why this was the case though?

[BBM]

I'm not sure if this is still the case, but if memory serves me correct, when the Thames Trains system was installed about a decade ago, there was a 3-minute threshold built in to the system for some reason. This meant that a delay of one, or two minutes would never be shown. A train would either be on time or 3+ minutes late. Not sure why this was the case though?

I think that's correct except that if a train which is 3 minutes late becomes 2 minutes late it's shown as 2 minutes late! But if it then becomes 1 minute late it's shown as 'on time'.

[Chris from Nailsea]

At Nailsea & Backwell, the CIS^▸ (Customer Information System) and public address system triggers any delay as '6 minutes', or '12 minutes' - in multiples of 6, certainly.  I suspect this is dictated by the track circuits between Yatton and Nailsea, which can't differentiate any smaller delays.  Quite often, having heard 'Digital Doris' announce a delay of 'six minutes', the train appears - just two or three minutes late!   

... and by the way, for details of the hymn writer (just in case anyone else is as sad as me, and is actually rather interested: see http://www.cyberhymnal.org/htm/n/e/newevery.htm   

[Hafren]

The ideal solution might be for the system to show 'Delayed' instead of an 'expected' time if the train is more than a few minutes overdue at a timing point. I think I've seen or heard of some systems doing that, but not sure off top of my head.

The very ideal situation would be for such trains to be flagged up somewhere so that a human can make an announcement based on what's really going on, as hopefully someone somewhere would know if a train has stopped somewhere (and whether/when it's started moving) or just not been recorded passing a timing point (e.g. some sort of signalling failure or re-routing). If the problem is affecting several trains then a human should be making at least some token announcements that there is a problem. If there's a CIS^▸ (Customer Information System) but no staff/PA^▸ (Public Address), maybe 'Delayed' could also include a message encouraging passengers to use the help point if there is one. Might be a bit of a strain on manpower, but it can be done sometimes, and would be 'good customer service'...

[onthecushions]

Dirty, Leaky Reading Station

Am I being too fastidious/picky in finding Reading (General) Station, the jewel in FGW^▸ (First Great Western)'s crown, a maintenance disgrace?

While the floors are swept and surfaces in reach are wiped, a glance above head height, particularly in the newer post-1991 parts, is stomach-churning. Much of Reading's spider population seems to have migrated there and it is evident from the settled dust, spiders' webs, pidgeon parts etc that 360 degree cleaning has never been done in 17 years of high-income generation. Some of this cleaning would only need the application of Ken Dodd's tickling stick, some would need an overnight access tower and a mop. Could one imagine an air-terminal with an annual footfall of 15M in this rank, dank condition?

Rain seems another unplanned-for eventuality (in England!). The roof drains on the modern large footbridge, to name just one site, consistently leak, leading to H&S^▸ (Health and Safety) "slippery floor" signs etc. I'm sure BR^▸ (British Rail(ways))/HMTreasury will be blamed for the original design (The GWR^▸ (Great Western Railway) used sloping rather than flat roofs, note) but all that expensive privatised post BR operation has not addressed these simple faults. At least BR was cheap.

Last night (November 3) the 1815 down ex-Paddington waited 14 minutes for a platform outside of Reading and we have been promised over ^400M to put this right. Enough of a buget for a mop, duster and the occasional person with a ladder, surely.

OTC

[dog box]

why do you think Reading would ever be regarded as the jewel in anyones crown???

[Super Guard]

The ideal solution might be for the system to show 'Delayed' instead of an 'expected' time if the train is more than a few minutes overdue at a timing point.

It does. Well at EXD^» (Exeter St Davids - next trains), if a waypoint is not triggered within 5+ minutes of its expected time it will flash "Delayed", and then update the expected time when it goes through the next station.  Of course a screen flashing "Delayed" begs questions from passengers, "Well how long is it delayed?" so from the front-line perspective that can be a bigger pain!

Castle Cary area seems to have problems, and the PAD^» (Paddington (London) - next trains)-PNZ trains can be flashing delayed because the waypoint has not triggered and all of a sudden the screen will go back to 'On Time' when the train arrives at Taunton, so in either way it is not ideal.

[Btline]

The new CT/LM^▸ (London Midland - recent franchise) screens are worse.

*They never record small delays (the most common).

*Often your train just disappears off the screen while you are waiting - a few minutes after it is due.

*If a incoming terminating train is late, it never seems to understand that the "new" service going out will also be delayed!

They have also done a platform alteration at WOF which I knew was not possible due to the track layout!

I stayed put as the crowds massed over to the other platform. A few minutes later, the screens went blank (and said "call 0890....etc. for info").

After a few moments the train came in on my platform - you can image the panic/anger that occurred when people realised they had been given false information! I wondered how it was able to give impossible info, and why the staff didn't realise it was impossible, CHANGE the screen (not blank it) and give an announcement.

Then again, the amount of times Digital Doris is cut off mid-sentence at Worcester is absurd (because she is talking absolute trash)!

[IndustryInsider]

The new CT/LM^▸ (London Midland - recent franchise) screens are worse.

The signalling systems in most of the London Midland station areas are what cause the problem. With Absolute Block signalling it is much more difficult to provide the information to allow the systems to work properly. The Cotswold Line stations suffer badly because of the same reason and regular travellers are sceptical at best when viewing the information shown on the screens.