Is the section of gwml beteen Paddington and reading the worst maintained line in the country?

[DailyCommuter]

(I was curious whether the overhead and side displays would have Tilehurst etc in their database(s). I can confirm they did! I filmed that too, of course…)

In previous timetables there have been daily scheduled IETs serving the stations between Didcot and Reading.

 

[PG]

South Today just had a report by Paul Clifton where he referenced a FOI query for disruption between Paddington and Reading in the year to November 2023. Apparently there were 361 days where there was some sort of disruption for which Network Rail were responsible. (And 363 days where GWR were).

@fgwrich has posted the link to that FOI upthread (post #289) in case you wish to delve further into the figures.

 

[Bald Rick]

South Today just had a report by Paul Clifton where he referenced a FOI query for disruption between Paddington and Reading in the year to November 2023. Apparently there were 361 days where there was some sort of disruption for which Network Rail were responsible. (And 363 days where GWR were).

I’d be bery surprised if there is *any* 30 mile stretch of 4 track main line in the country with similar levels of use as the GWML that does not have some sort of delay incident on it every single day. ’Disruption’ can be anything from all the wires down delaying thousands of people for hours, to an error in the day’s Automatic Route Setting data file that delays one train by 3 minutes.

 

[AngusH]

Following from the FOI request data.

Delays and Issues on GWML between Reading and London Paddington - a Freedom of Information request to Network Rail Limited

Please can you provide the following statistics relating to all lines (relief and fast), between Reading and London Paddington: Number of days containing any disruption caused by: 1) Signal Failure 2) Points Failure 3) Overhead Wire Damage 4) Broken or Damaged Rail 5) Industrial Action 6)...

www.whatdotheyknow.com

with the spreadsheet (html+link)

FOI202300378 Data.xlsx

www.whatdotheyknow.com

I did some quick graphs of the values and it becomes clear that there are some good improvements in number of days affected by signals and overhead wire issues.
Broken rails and points seem worse though.

I didn't bother with the El/GWR split, the EL situation is always fewer affected days than GW so far as I could see.

The graph image is generated from data in the FOI spreadsheet provided and may be read there if the image is not accessible.

Any errors in graphing are my own and will be corrected if advised.

 

[Sly Old Fox]

A delay caused by the operator could be one train being held up for 30 seconds whilst a ramp is provided for a disabled passenger and suddenly the operator is responsible for that day having a delay marked against it. Whereas infrastructure problems tend to affect a lot more.

 

[Bald Rick]

A delay caused by the operator could be one train being held up for 30 seconds whilst a ramp is provided for a disabled passenger and suddenly the operator is responsible for that day having a delay marked against it. Whereas infrastructure problems tend to affect a lot more.

‘disruption’ will be for any delay registered in TRUST (the system that records train running and delays) occurring in that section.

Usually this requires a delay of 3 minutes or more, except for temporary speed restrictions where delays of one minute or more can be recorded.

 

[Horizon22]

South Today just had a report by Paul Clifton where he referenced a FOI query for disruption between Paddington and Reading in the year to November 2023. Apparently there were 361 days where there was some sort of disruption for which Network Rail were responsible. (And 363 days where GWR were).

361 days of disruption I imagine means just 1 cancellation, alteration or delay. I’d be shocked if there’s any day where there isn’t and probably Christmas Day and Boxing Day are 2 of them!

Following from the FOI request data.

Delays and Issues on GWML between Reading and London Paddington - a Freedom of Information request to Network Rail Limited

Please can you provide the following statistics relating to all lines (relief and fast), between Reading and London Paddington: Number of days containing any disruption caused by: 1) Signal Failure 2) Points Failure 3) Overhead Wire Damage 4) Broken or Damaged Rail 5) Industrial Action 6)...

www.whatdotheyknow.com

with the spreadsheet (html+link)

FOI202300378 Data.xlsx

www.whatdotheyknow.com

I did some quick graphs of the values and it becomes clear that there are some good improvements in number of days affected by signals and overhead wire issues.
Broken rails and points seem worse though.

I didn't bother with the El/GWR split, the EL situation is always fewer affected days than GW so far as I could see.

The graph image is generated from data in the FOI spreadsheet provided and may be read there if the image is not accessible.

Any errors in graphing are my own and will be corrected if advised.

View attachment 150631View attachment 150635

Seems to me it's generally physically things wrong with the rails themselves - this has certainly manifested itself in cracked crossings and broken rails the last 6 months.

 

[ChewChewTrain]

In previous timetables there have been daily scheduled IETs serving the stations between Didcot and Reading.

I’d be interested to see such a timetable. I knew HSTs had served those stations in the peaks for a few years, but thought the IETs that replaced them never did*, the new 377s having solved the peak-time capacity issues there were with the Turbos.

—
* with the possible exception of the weird early-morning Sunday service that reappears every few years and has an inexplicable DID-GOR-RDG stopping pattern.

 

[DailyCommuter]

I’d be interested to see such a timetable. I knew HSTs had served those stations in the peaks for a few years, but thought the IETs that replaced them never did*, the new 377s having solved the peak-time capacity issues there were with the Turbos.

From the 20th May 2019 there was a service which arrived into Reading around 06:45, starting from Oxford which was previously a 6 car turbo and became a 5 car IET, although was a 9 car did show up on some days. From Dec 2019 the service was curtailed to start at Didcot and became a 12 car 387 providing much needed capacity for the Twyford and Maidenhead passengers, as it was cosy on a 5 car IET.
There was also an IET in the evening peak serving the stations between Reading and Didcot and continued onto Banbury.
The HST service started at Didcot, arriving Reading around 07:05 and was initially replaced with a 6 car turbo, then become 387s at electrification.

 

[Snow1964]

Am I reading the data in #304 correctly

Basically 363 days out of 365 was a delay !

 

[takno]

Am I reading the data in #304 correctly

Basically 363 days out of 365 was a delay !

It's not really surprising on a busy bit of railway that at least one train would have been delayed by at least 3 minutes on every operating day, which is why it's not a very useful data set to collect. The FOI response really should have attempted to provide some context on how minor this could be.

Even the breakdown doesn't provide much context on whether for example there was a major signalling failure or a single signalling unit failed to register correctly once and had to be reset.

 

[Annetts key]

"Signalling failure" is also a VERY wide and extensive category. Many people outside the S&T and signallers (and some operations staff) are unlikely to truly understand.

 

[Class 800]

Is this the new OLE that is coming down or the PAD-Heathrow Jn. headspans installed pre-HEX?

 

[devon_belle]

It would be interesting to see failure data normalised by the number of trains.

Also, I'd be interested to see whether all these failures unique or are there repeat locations/equipment?

 

[Bald Rick]

yes, on 363 days of the year, at least one train was delayed by 3 minutes or more between Paddington and Reading.

In other news, around 1,000 trains a day run on the section between Paddinton and Reading inclusive.

 

[Annetts key]

In other news, around 1,000 trains a day run on the section between Paddinton and Reading inclusive.

I would have thought that you would know or be able to find out the exact number of timetabled trains...

 

[Bald Rick]

I would have thought that you would know or be able to find out the exact number of timetabled trains...

I could - as could anyone - but what would be the point. Besides, it varies every day!

 

[oxfordrelayer]

Maintenance staff have been cut, it’s not a surprise. It’s the future.

You got it spot on

 

[Edward Betts]

I recently contacted Liam Fox, my local MP, to express my concerns about the line's maintenance and performance issues. His office forwarded these concerns to GWR and I've received a response from them.

From: Thomas Lydon
Sent: 19 January 2024 16:39
To: Office of The Rt Hon Dr Liam Fox MP
Subject: RE: Urgent Need for Improvements in Train Services on the Great Western Main Line

Dear Sir Liam

Thank you for your email on behalf of your constituent, Mr Betts.

He is right, performance has not been good enough recently. Last year and the beginning of this year have been particularly difficult, with the ongoing industrial action (with more to come at the end of this month and beginning of February unfortunately) and an increasing number of weather-related disruptions (including flooding between Bristol and Nailsea), which are not easy to manage, but there have also been too many issues relating to the infrastructure (such as cracked rails, points and signalling problems, and issues with the overhead electric lines), which is managed by Network Rail – particularly between Reading and London Paddington which affects all of our high-speed services including Bristol, Nailsea & Backwell, and beyond.

You may be aware that on 29 November the Office of Rail and Road (ORR) launched an investigation into poor train punctuality and reliability in the Network Rail Wales & Western region, and Network Rail have committed to work with the ORR to identify causes and take steps to address. We have also launched a project which seeks to stable the current situation with the infrastructure, and our senior management team meet with Network Rail each week to review their progress on this.

We did have a number of cancellations because of more IET sets being out of service than normal. This was due to significant damage, and a number that have required heavy maintenance all at the same time, which should have been staggered. We are working with Hitachi Rail Europe (HRE) to resolve this, and we have also used an independent consultant to review Hitachi’s maintenance plan against ours, to form a better plan and to avoid too many trains being unavailable at the same time.

Some of the things impacting our service are not quick fixes, but I want to assure you that our performance and reliability is our priority, and we are working very hard with our industry partners to improve this. Mr Betts mentions Delay Repay, if he does need any assistance in claiming I would be happy to help. Like him, while compensation is an important part of recognising when things haven’t gone right, we would much rather our services ran to time.

I hope this helps and thank you again for writing.

Best wishes

Tom

Thomas Lydon | Public Affairs Manager | Great Western Railway
Milford House | 1 Milford Street | Swindon | SN1 1HL

 

[Jamiescott1]

I recently contacted Liam Fox, my local MP, to express my concerns about the line's maintenance and performance issues. His office forwarded these concerns to GWR and I've received a response from them.

I also contacted my mp and recieved a similar response from Huw Merrimen- secretary of state for transport

 

[GRALISTAIR]

It is easy to “talk the talk”. Time to “walk the walk” me thinks.

 

[Horizon22]

It is easy to “talk the talk”. Time to “walk the walk” me thinks.

Who are you referring to here? Both Network Rail's directors for the area have been in place less than a year.

 

[Bald Rick]

It is easy to “talk the talk”. Time to “walk the walk” me thinks.

It’s fair to say that performance of the railway is like a supertanker - it takes a long time from inputting commands to the thing turning in the right direction. It was ever thus.

I have been personally involved in several similar situations; it takes about 6 months before there are noticeable improvements. In this case I reckon we’re about half way through that.

 

[Adrian1980uk]

It’s fair to say that performance of the railway is like a supertanker - it takes a long time from inputting commands to the thing turning in the right direction. It was ever thus.

I have been personally involved in several similar situations; it takes about 6 months before there are noticeable improvements. In this case I reckon we’re about half way through that.

Well with infrastructure, you can change/replace ohle or track etc. next week, it's takes months of planning

 

[GRALISTAIR]

It’s fair to say that performance of the railway is like a supertanker - it takes a long time from inputting commands to the thing turning in the right direction. It was ever thus.

I have been personally involved in several similar situations; it takes about 6 months before there are noticeable improvements. In this case I reckon we’re about half way through that.

That is good to hear.

Who are you referring to here? Both Network Rail's directors for the area have been in place less than a year.

I am not assigning blame but dread the thought of a major accident on the stretch of line.

 

[Bald Rick]

Well with infrastructure, you can change/replace ohle or track etc. next week, it's takes months of planning

But that is the point - you can replace some thigns tomorrow, but how do you know that will make a difference? You have to assess what is wrong, identify the trends, find the root causes, plan in the work, deliver it, and then

Taking the spate of broken rails for example - there’s about 300 miles of rail between Paddington and Reading. Obviously you will prioritise replacement of areas with defects in, and the worst defects first, but the laws of Messrs Murphy and Sodde determine that some defects will worse. much more quickly than anticipated and a relatively minor one will fail the week before you were due to replace it (as has happened).

 

[brad465]

The latest RAIL edition has a small section titled "Faults disrupt passengers GWML every day!" with a summary of the problems including 200 days with points failures, 175 days for broken rail, 97 days for signal failures and 90 days for damaged OHLE. It does say though in the year "to November 30 2022".

 

[PG]

The latest RAIL edition has a small section titled "Faults disrupt passengers GWML every day!" with a summary of the problems including 200 days with points failures, 175 days for broken rail, 97 days for signal failures and 90 days for damaged OHLE. It does say though in the year "to November 30 2022".

They have obviously seen the FOI request referenced upthread but not bothered about proof reading!

 

[Purple Train]

They have obviously seen the FOI request referenced upthread but not bothered about proof reading!

There has been a noticeable downgrade in the quality of proof-reading since much of the editorial team left.

yes, on 363 days of the year, at least one train was delayed by 3 minutes or more between Paddington and Reading.

In other news, around 1,000 trains a day run on the section between Paddinton and Reading inclusive.

Would those two days be Christmas Day and Boxing Day, or am I being thick?

 

[Horizon22]

Would those two days be Christmas Day and Boxing Day, or am I being thick?

No you’re not, that is quite likely.