Boris demands TfL agree to Driverless Trains part 2

[spsmiler]

Driverless trains have been around for decades - but apart from perhaps the Nuremberg Germany U-Bahn plus parts of the Paris metro this has always been on new-build services designed for driverless trains.

Here in London we did have the FACT Project on the Hainault - Woodford route (FACT = Fully Automatic Control of Trains)

I understand this as having been concluded in 1978 - due to lack of funds.

This website may be of interest, I admit it is some years since I reviewed its information.

metroautomation.org/ (Observatory Of Automated Metros)

[aslefshrugged]

Nuremburg U-2 was the first to be converted from driver operated to driverless 2009-10 but its only 8.2 miles long with 16 stations, U-3 was built driverless and shares track with U-2. There are currently no plans to convert U-1 to driverless.

Paris M 1 was converted 2007-2012, 10.3 miles serving 25 stations, all platforms were fitted with platform edge barriers. Management negotiated a deal with the drivers union CDG so that drivers were gradually displaced to other lines as driverless trains replaced driver operated trains. In 2013 they announced plans to automate M 4, 7.5 miles long with 27 stations.  Siemens were awarded the contract in 2016, the first driverless trains were introduced in 2020 and it should go fully automated next year

The only other lines to be converted to driverless are North South MRT and East West MRT in Singapore (2017-19) but they had the advantage of already having platform edge barriers (installed in 2011-12)

All other driverless metros were built driverless including Paris M 14.

[brigham]

I hadn't realised it was so widespread.

When I first saw this thread, I was led to believe that Johnson had taken leave of his senses.

Turns out, he's only wanting what other cities already have!

[Chris M]

No, Boris doesn't want driverless trains because that's what other cities have, he wants them because he lost the battle to bring them in when he was mayor of London. He wanted them then because drivers are expensive and go on strike. That driverless trains are actually even more susceptible to strike disruption and would cost more are facts that don't accord with ideology and so don't matter to him.

[North End]

Sept 29, 2021 17:37:04 GMT Chris M said:

No, Boris doesn't want driverless trains because that's what other cities have, he wants them because he lost the battle to bring them in when he was mayor of London. He wanted them then because drivers are expensive and go on strike. That driverless trains are actually even more susceptible to strike disruption and would cost more are facts that don't accord with ideology and so don't matter to him.

This is exactly it - it’s ideological.

If you or I were building LU today, we’d probably want to go with unmanned operation. The benefits in terms of cost and operating simplicity are very attractive - indeed we saw elements of this during the Covid lockdown special services when lines were able to recover from lengthy suspensions in the blink of an eye, the difficulty being this is only possible when you are only running a handful of trains.

In the real world, LU as it exists simply isn’t built for unmanned operation. The physical characteristics simply aren’t setup for it, and the only way to rectify that would be to essentially rebuild the whole system from scratch. The stations are a long way apart in many places, most of the system has no smoke control measures, and there is no dedicated emergency access in the tunnels. Even relatively recent stuff like DLR to Bank or the JLE has design issues - in particular the trackside walkways are too narrow to avoid a conflict between escaping passengers and arriving emergency services staff. I think I’m right in saying that there’s only *one* part of LU whose design adequately provides for this, and that’s the Northern Line to Battersea. Perhaps Heathrow Terminal 5 as well, I can’t remember. These factors essentially exclude unmanned operation, at least in the tunnel sections.

Now you have a human on the train, you’re back to the territory of duty schedules, the abolition of which is probably the single biggest prize of trying to eliminate the traditional train operator, simply because it makes operation more flexible. There’s also an argument to say that if you’re going to have a member of staff on the train, a cab is the optimum place for them to be located, even if their role might be more focussed on monitoring CCTV than observing the road ahead. Covid only reinforces this - look at the issues operators have had where there has not been a cab for staff to work from, we’ve seen all sorts of abominations like areas having to be cordoned off, or in some cases entire carriages being designated for staff use. Not having a place of safety for a member of staff is poor practice, and DLR only get away with it because “that’s how it is” on a system built as a bit of a toy railway, one of a number of features which probably wouldn’t be acceptable if one were designing the system from scratch today.

The following is a personal point not a party political one. The crusade for abolishing the traditional train operator flows from the ideology of one individual - Alexander Boris Johnson. He has form for having a personal grudge against train drivers, I remember coming across an article of his when he was simply lowly editor of the Spectator, slagging off the driver of the turbo at Ladbroke Grove for “ignoring three danger signals”. At the time I found this at best ignorant and poorly informed, and at worst crass and insensitive, when there has never been any suggestion the Ladbroke Grove driver “ignored” the warning aspects. As we all know, the subsequent investigations uncovered numerous deficiencies in both the infrastructure and the driver’s training, so Johnson got that one very wrong indeed.

As for where this will go, this same individual has presided over a period where well over 100,000 people have lost their lives. Notwithstanding whether things could have been done differently or not, I’m not sure any leader could or should expect to remain in power long-term after such an event. As a country we are going to have to move on from Covid, and the events of the past week have perhaps highlighted this. Fresh leadership is bound to be sought as part of the moving-on process. I can’t see this individual lasting in power too long, and with his demise I predict the fad for driverless trains will go the same way. We just have to hope not too much money and time gets wasted on it in the meantime, or we don’t get lumbered with something like a fleet of trains which are then unfit for purpose for 40 years thanks to one man’s pet ideology.

In the meantime, at the moment LU has a list of issues as long as your arm, and the distraction of pet projects is extremely unhelpful all round.

[aslefshrugged]

Sept 29, 2021 16:47:09 GMT brigham said:

I hadn't realised it was so widespread.

When I first saw this thread, I was led to believe that Johnson had taken leave of his senses.

Turns out, he's only wanting what other cities already have!

There are far more driver operated metros than there are driverless and conversion to driverless is rare (4 lines in three cities, soon to be 6 lines in four cities).

---

I forgot one conversion, in 2011 the Scottish government announced that the  "Clockwork Orange" in Glasgow would go driverless, in 2016 the contract for new trains and signals was awarded to a consortium of  Ansaldo/Stadler.

The first train was supposed to enter service last year but there has been delays due to problems fitting the new signal system in Victorian tunnels.  The first trains will enter service next year on the old signals, they will be driver operated until all the old stock has been withdrawn and platform edge doors have been installed

[jimbo]

The current DLR trains order remain cabless as current DLR fleet despite problems through Covid. here They could be fitted with temp cab screen as new Picc trains will be.

[brigham]

Sept 29, 2021 17:37:04 GMT Chris M said:

No, Boris doesn't want driverless trains because that's what other cities have, he wants them because he lost the battle to bring them in when he was mayor of London. He wanted them then because drivers are expensive and go on strike. That driverless trains are actually even more susceptible to strike disruption and would cost more are facts that don't accord with ideology and so don't matter to him.

The point I was making is this: 

Driverless trains on urban railways similar to the Underground DO exist, and are growing, not shrinking.
My impression was that Johnson was on a flight of fancy; in reality, he's just incapable of accepting the limits of the existing technology as far as London is concerned.

The same with the 'death ray' argument. He was seen as a 'nut' because the existence of these weapons wasn't widely known. It turned out that he was talking about an established weapons system, and I was the one who was ignorant.

Johnson is no 'flat-earther'. He just attracts the same publicity.

[Chris L]

In this thread, brigham wrote:

If driverless trains are the 'in thing' in foreign parts, it might be an idea to keep an eye on them; although I don't think 'keeping up with the joneses' is a good maxim for railway development.

There are big differences between new systems around the World built for driverless trains and the Underground, particularly the deep level tunnels.

Safe walkways for evacuating passengers in dark tunnels do not exist except on extensions and Crossrail.

I think it would be a very good idea to take ministers and DfT seniors on a few tunnel walks and see what they think afterwards.

Posts merged from Covid implications thread which was drifting - Tom

[d7666]

Feb 28, 2022 11:17:04 GMT Chris L said:

Feb 28, 2022 9:01:24 GMT brigham said:

If driverless trains are the 'in thing' in foreign parts, it might be an idea to keep an eye on them; although I don't think 'keeping up with the joneses' is a good maxim for railway development.

There are big differences between new systems around the World built for driverless trains and the Underground, particularly the deep level tunnels.

Safe walkways for evacuating passengers in dark tunnels do not exist except on extensions and Crossrail.

I think it would be a very good idea to take ministers and DfT seniors on a few tunnel walks and see what they think afterwards.

I can't think of one single heavy metro line anywhere in the world that is driverless that is NOT :

[1] a new construction purpose built and opened with driverless operation
[2] new construction purpose built for driverless, not driverless at opening, subsequently driverless
[3] m-a-j-o-r rebuild of an existing line so completely rebuilt effectively a new build bearing no resemblence to the previous line except geographical line of route

I am sure there might be an exception to that somewhere on the planet and equally sure my post will catalyse the District Dave collective into posting a rebuff. Either way, it won't be a significant proportion of the world's heavy metros.

Some time, years, ago, we got dragged into a presentation at one of the large hq offices where the main presenter repeatedly compared London with the Sao Paulo [Brasil] metro.

I could only conclude [1] the presenter had never set foot in the place [2] picked Sao Paulo in the hope that no-one in the room would have any knowledge of it so not question the slides.

I had to keep quiet. I did not know where to start.

I spent 3 years living in Sao Paulo of which 18 months involved commuting daily on the metro and the rest of the time routinely using it for other travel. Sao Paulo metro is all ATO right from outset, the oldest route dates from 1974. At the time I was there it was the first 3 core lines in a simple star network. I'd visited their command centre, and one of the depots, both on technical professional visits.

London underground and Sao Paulo metro are chalk and cheese infrastructure - which was omitted by the presenter - but so are the state politics, the public transport finances, the country economy, the country inflation, the urban geography, the demographics, virtually flat traffic levels from start to finish, no real peaks, and it serves only the inner area in a city of three times the population of greater London - but compacted into an area about three times smaller.

This is the problem. Looking somewhere else in the world, selectively picking out certain data, prepping a slide from it, and making that policy.

The same applies to driverless trains.

This is some digression from coronavirus subject thread though ................

[Dstock7080]

Feb 28, 2022 15:05:04 GMT d7666 said:

I can't think of one single heavy metro line anywhere in the world that is driverless that is NOT :

[1] a new construction purpose built and opened with driverless operation
[2] new construction purpose built for driverless, not driverless at opening, subsequently driverless
[3] m-a-j-o-r rebuild of an existing line so completely rebuilt effectively a new build bearing no resemblence to the previous line except geographical line of route

I am sure there might be an exception to that somewhere on the planet and equally sure my post will catalyse the District Dave collective into posting a rebuff. Either way, it won't be a significant proportion of the world's heavy metros.

This is some digression from coronavirus subject thread though ................

not as a rebuff but an example you're looking for is Paris Metro Line 1.

(Last line removed as it's now out of context following post move - Tom)

[d7666]

I perhaps ought to have added a suggestion a mod shunt my post to another thread if drifting too much, but before I got back on line again to do this, you got there first.

Always a hard one to know what to do, when someone does begin a drift, but that drift needs further drifting response and we have a Mary Celeste.

PS I didn't know Paris U1 was driverless now, I thought it was only ATO. I plead not been to Paris for about 8 years now.

[aslefshrugged]

Feb 28, 2022 16:18:46 GMT Dstock7080 said:

Feb 28, 2022 15:05:04 GMT d7666 said:

I can't think of one single heavy metro line anywhere in the world that is driverless that is NOT :

[1] a new construction purpose built and opened with driverless operation
[2] new construction purpose built for driverless, not driverless at opening, subsequently driverless
[3] m-a-j-o-r rebuild of an existing line so completely rebuilt effectively a new build bearing no resemblence to the previous line except geographical line of route

I am sure there might be an exception to that somewhere on the planet and equally sure my post will catalyse the District Dave collective into posting a rebuff. Either way, it won't be a significant proportion of the world's heavy metros.

This is some digression from coronavirus subject thread though ................

not as a rebuff but an example you're looking for is Paris Metro Line 1.

There is a dedicated thread about driverless trains, and you are right, this isn't the thread for that.

So far four lines have been converted from driver operated to driverless

2010 Nuremburg U2.  Shared track over the central section with driverless U3 (opened 2008)

2011 Paris Metro 1.  Took four years to convert.  New trains (MP05) had same door configuration as old trains (MP89), trains were driver operated while Platform Edge Doors were installed  Prior to conversion RATP negotiated a deal with CGT drivers union, all drivers were displaced to other lines after conversion was complete (with hefty bonuses all round)

2017 Singapore North South
2018 Singapore East West.  Both lines already had Platform Edge Doors

Currently mid-conversion

2023 Paris Metro 4.  Five year conversion, I imagine they're following the same pattern as M1

Possible future conversion

Glasgow Subway,  Will have PEDs.  At the moment the new trains are undergoing trial running but there have been delays installing the new signal system into the old Victorian tunnels

All other driverless railways were built driverless and all other driver operated railways are still driver operated

TfL and the government can consdier driverless operation on the Tube as much as it wants, it would be expensive, take decades and would need the unions co-operation.  A red herring

[rsdworker]

well yeah why not look at glasgow subway - that has no sudewalks in tunnels its possible evacuate via front or rear doors at ends - simllar to singpore where ramps installed to allow evucation to track

[Tom]

You're right there are no sidewalks, but they also have only a third rail and very few points and crossings, resulting in far fewer trip hazards.

[jimbo]

Glasgow has shorter trains, and much lower passenger numbers per train. In the event of a train stalled between stations, the numbers to be evacuated would be much smaller.

[class411]

Mar 1, 2022 8:54:50 GMT aslefshrugged said:

TfL and the government can consider driverless operation on the Tube as much as it wants, it would be expensive, take decades and would need the unions co-operation.  A red herring

Yes, it would cost billions* (literally) and, even if TFL ever have that much budget available, will there not always be something more important on which to spend it?

*Presumably, the capability of the system will get nearer to that required for fully automatic operation as trains and signalling are upgraded, but it's always going to be a massive money sink.

[aslefshrugged]

TfL have postponed a whole load of planned works due to lack of funds

Piccadilly signals upgrade
Bakerloo new trains, signals upgrade and extension
Camden Town, Elephant & Castle and Holborn rebuilds
DLR extension to Thamesmead

Plus the Rotherhithe tunnel is in need of some serious work and could close at some point

TfL have been told to look at driverless operation on the Piccadilly and Waterloo & City, Piccadilly would be a nightmare but Waterloo & City would be comparatively simple; shut down for however long it takes, new trains, signals upgrade and install PEDs on the platforms.

Sadly that wouldn't happen before I retired so I'll have to carry on working down the Drain at least once every six months to keep my licence up to date!

[jimbo]

Mar 2, 2022 0:00:51 GMT jimbo said:

Glasgow has shorter trains, and much lower passenger numbers per train. In the event of a train stalled between stations, the numbers to be evacuated would be much smaller.

This is a reason to pick the Waterloo & City line as a demo line in London. Boris is still pushing for driverless trains, even in the latest funding agreement. If he wants a line converted to show that London can do what Paris can do, this is the one. We know it's not a typical line, with only five trains and four platforms, but most people will have never used it. It need not involve TfL if the line is taken from them, just as it was given to them some time back. A group could be contracted to close and convert the line with new trains and signals, and then operate it for say 25 years.

[class411]

Mar 2, 2022 19:10:17 GMT jimbo said:

Mar 2, 2022 0:00:51 GMT jimbo said:

Glasgow has shorter trains, and much lower passenger numbers per train. In the event of a train stalled between stations, the numbers to be evacuated would be much smaller.

This is a reason to pick the Waterloo & City line as a demo line in London. Boris is still pushing for driverless trains, even in the latest funding agreement. If he wants a line converted to show that London can do what Paris can do, this is the one. We know it's not a typical line, with only five trains and four platforms, but most people will have never used it. It need not involve TfL if the line is taken from them, just as it was given to them some time back. A group could be contracted to close and convert the line with new trains and signals, and then operate it for say 25 years.

The thing is, though, what will this achieve?

Show that a point to point line with two stations can operate fully automatically?

Gatwick's been doing that for years. Perugia (and many others) have also been doing a lot more for years.

Show that an extant line can be converted to fully automatic operation?

Far more ambitious schemes have been implemented before. All it's likely to do is annoy a lot of highly paid bankers and prove virtually nothing that has not already been demonstrated.

It would make far more sense to do an end of line segment (e.g. Hammersmith to Royal Oak) that could be implemented with virtually no disruption. You could probably do that by 2030, and have the whole network converted by (depending on available funding) 2060.

[1018509]

Aug 21, 2021 22:19:38 GMT class411 said:

Aug 21, 2021 21:16:27 GMT jimbo said:

This is part of the funding conditions for the last short-term grant deal, and was part of the previous one also!

Indeed.

If they want the money they have to agree that at some point they will form a task group that will investigate the setting up of a working party that would define the broad terms of an inquiry into the feasibility of instigating a commission to consider whether of not to recommend the drafting of a report on the possibility of introducing driverless trains at some point in the future.

My guess is that they will agree.

Surely they would talk about this first.

[brigham]

With hindsight, it might have been better for the W & C to have been made driverless during the last refurbishment.

How soon AFTER the decision to close and refurb it did the 'driverless' notion appear?

[toby]

Is the notion the engineer-side confidence in technical ability to retrofit this specific line, or the need for the political attack?

I think there was something optional considered for the last refurbishment, but not done because it would not be a good use of an extended shutdown.

[aslefshrugged]

Mar 3, 2022 8:25:43 GMT brigham said:

With hindsight, it might have been better for the W & C to have been made driverless during the last refurbishment.

How soon AFTER the decision to close and refurb it did the 'driverless' notion appear?

W&C would need a signal upgrade to go driverless, the stuff we work on is left over from British Rail.  You'd also need new rolling stock as I doubt very much the Class 482/1992 Tube Stock could be retrofitted for a new CBTC signalling system or to work with Platform Screen Doors.  The Central Line signalling system and ATO is simply too unreliable to be used for driverless operation (90s technology)

TfL investigated the idea of driverless operation on the Piccadilly, Bakerloo, Central and Waterloo & City Lines in the New Tube for London Feasibility Report October 2014 although it concluded it wouldn't happen on the Bakerloo due to sharing track north of Queens Park.  The new trains for the Waterloo & City Line would have been delivered in 2032 but now nobody knows

[johnlinford]

Surely all they need is a short document that says "to upgrade to 'driverless' will cost X, cause Y disruption and save Z in anual running costs for reasons a, b and c. It is therefore uneconomical. If the DfT would like to proceed, please sign a blank cheque."

[Chris M]

The response to that seems to have been the requirement to research technologies that will make them more affordable.

[class411]

Mar 3, 2022 14:28:34 GMT Chris M said:

The response to that seems to have been the requirement to research technologies that will make them more affordable.

Which, in terms of timescale, is extremely open ended.

[johnlinford]

Indeed. I could see it hypothetically working, if you build Crossrail 2 (and 3) and use the capacity from that to close certain lines for extended periods for major refit to allow for it. Oh look, we're back to DfT writing blank cheques again.

(Of course it won't stop strikes because there will still be control room staff and it won't save a huge amount in running costs as a result).

[aslefshrugged]

Mar 3, 2022 15:32:53 GMT johnlinford said:

Indeed. I could see it hypothetically working, if you build Crossrail 2 (and 3) and use the capacity from that to close certain lines for extended periods for major refit to allow for it. Oh look, we're back to DfT writing blank cheques again.

(Of course it won't stop strikes because there will still be control room staff and it won't save a huge amount in running costs as a result).

Control Room staff, train maintainers, track workers, signal technicians, power supply technicians, station staff and DLR style on-board staff.  Most of which will be RMT, all you do by removing the drivers is remove ASLEF from the equation

The DLR PSAs are on about £10k less than Train Operators

[d7666]

Mar 2, 2022 22:40:58 GMT class411 said:

The thing is, though, what will this achieve?

Show that a point to point line with two stations can operate fully automatically?

Gatwick's been doing that for years. Perugia (and many others) have also been doing a lot more for years.

Gatwick may be point to point but is pointless. That is, no points a.k.a. no switches and crossings.

Far simpler set-up, orders of magnitude simpler. Two completely segregated guideways, one train dedicated train to each guideway.
One train can't possibly infringe the other's movement authority except in certain derailment (deguidewayment?) scenarios but that equally applies anything driverless or crewed so a specious issue.

You may as well cite Bournemouth or Hastings etc cliff lifts as examples of driverless operation for all the relevance Gatwick has to any London heavy metro line. Or let's go fit cable haulage around the Circle line.


But anyway you don’t need to demonstrate driverless operation or reprove it can be done. London had a driverless tube railway in 1930s it is known it can be done. Given the ten figure sums just to do 4LM CBTC, what needs to be shown is can driverless be done AND deliver the alleged cost savings when counting EVERYTHING needed be it PEDs and or new stock and or degraded mode working with staff trained in train movement and so on and on and on.