Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Oct 7, 2024 21:41:09 GMT
Given that converting the whole network to driverless trains in one go would take literally billions of pounds, take an absolute minimum of ten years (realistically probably more) given the need to design, build and test new rolling stock, design and install all the signalling, etc. and would actually make the system more vulnerable to strikes (you still need signallers, controllers, station staff, cleaners, etc; a line requires hundreds of drivers but only a handful of signallers and controllers).
I don't think anywhere close to 1% of the travelling public would support that plan even if it wouldn't crash London's (and possibly the country's) economy (which it would).
|
|
|
|
Post by sealingmark on Oct 8, 2024 7:16:35 GMT
Wonder if there's much quicker and cheaper way...
Rather than rebuild the entire system from the ground up, start with copying the technology from self-driving cars. That isn't perfect, but I believe it already kills fewer people per mile driven than human car drivers.
Each train would need to be equipped with cameras/lidar and each cab with a black box to interface to the existing train controls. Job done.
Signalling is interesting.
Say a decision is made to change signal XXX from G to R. Somewhere in a computer the value of a byte changes.
If that data can be conveyed to the train (not easy I realise), it becomes unnecessary for a message to be sent along milles of wire to cause a lamp to change colour. The entire signalling infrastructure becomes redundant.
Just a thought.
FWIW, I am not attacking train drivers, I'm criticising the system of industrial relations which results in the pain being inflicted on the travelling public, who aren't responsible and can't do anything about it.
|
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Oct 8, 2024 11:14:30 GMT
The issue is that, as far as I am aware, it is not possible to retrofit self-driving technology (whether rail or road based) to trains that were not designed for it. Even if it were, equipping every train with the technology you describe would still cost tens to hundreds of millions of pounds, minimum, for the entire LU fleet. And that is not counting the time and money required for testing the systems and getting all the necessary safety authorisations. Installing any new signalling system inevitably requires extensive closures of large sections of line, and as a rule of thumb the greater the change from the existing system the more disruptive the change is. Do also remember that there is a finite capacity for manufacture and installation/introduction and testing of new trains and signalling systems - even if you placed all the necessary orders tomorrow the project would not be finished within a decade.
As noted it will also not solve the issue of poor industrial relations as all the signallers, controllers, maintenance staff, station staff, and on-train staff, etc will still be union members (note how the automatically-driven DLR is not immune from disruption due to strike action). The fewer staff that are required to run a train service the greater the impact of any one of them going on strike. For example at present, ten drivers striking would result in delays at most*, but five line controllers striking would result in no service on significant parts of the network. Also don't forget that the staff involved in designing, manufacturing, installing, and testing all the new equipment for your automatic operation are also union members who can and will strike if you don't treat them well enough.
*10 drivers, all from the same small depot, could result in severe delays or a part suspension. 2 drivers at each of 5 different depots would likely not be noticeable to the travelling public.
It is indeed unfortunate that industrial relations issues impact the travelling public, but the way to solve this is to fix the industrial relations such that any disputes are resolved long before they get to the point of staff feeling the need to take strike action. Remember that staff do not like striking - it costs them money and as well as disrupting their family and friends in the same way that you are affected, and puts a lot more pressure on staff who are not striking too. Strikes are always a last resort.
While fixing industrial relations issues it isn't as simple as giving everybody more money, properly funding the transport network is a key aspect of it.
|
|
|
|
Post by sealingmark on Oct 8, 2024 11:52:21 GMT
Fixing industrial relations in the rail industry seems as likely as achieving Middle East peace.
So I guess Joe Public just has to put up with strikes indefinitely.
Just because automation is extremely difficult, doesn't mean it shouldn't start one day.
No doubt it would be sold to the unions by having 'drivers' in the cab monitoring the automatic operation, for a generation.
Hey, they could even work from home with a couple of work-supplied screens, if they chose to do so!
And a driver could earn extra by monitoring more than 1 train, if they wished.
Always a way of squaring the circle.
Old chestnut from the Airline industry; by 2030 [update year as required], the flight crew will comprise one man and one dog.
The duty of the man is to Not touch anything.
The duty of the dog is to bite the man if he fails to perform his duty :-)
|
|
|
|
Post by billbedford on Oct 8, 2024 14:54:25 GMT
FWIW, I am not attacking train drivers, I'm criticising the system of industrial relations which results in the pain being inflicted on the travelling public, who aren't responsible and can't do anything about it. Corse the travelling pulic are responsible, they are the ones who elected government, both national and local. |
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Oct 9, 2024 0:09:03 GMT
Just because automation is extremely difficult, doesn't mean it shouldn't start one day. Automation of the driving task on London Underground began in the 1960s on the Victoria line. Despite 60 years of investment and technological progress, drivers are still required. Why? Because running a safe railway with no staff onboard is an incredibly complicated task, made even more so by needing to interact with people in many locations. The obvious comparison is with airport people mover systems, but they are topographically extremely simple compared to even the simplest LU line (Waterloo & City), have lower usage and the track is completely segregated from passengers and other vehicles. No doubt it would be sold to the unions by having 'drivers' in the cab monitoring the automatic operation, for a generation. The reason drivers are in the cab on current automatically operated lines (Victoria, Central, Jubilee) is so they can take over in an emergency (e.g. a person on the track) or degraded conditions (e.g. signal failures), resolve many defects, and assist passengers when required. The DLR train attendants also act as revenue protection to a degree. All these staff are required for the safe operation of the railway and they can and do strike, so this doesn't achieve your goal. Hey, they could even work from home with a couple of work-supplied screens, if they chose to do so! So are you going to pay for the bandwidth and distraction-free work environment at every driver's home? What about the uninterruptible power supply? IT issues like the recent cyber attack (remote working staff couldn't (maybe still can't?) remotely access things like Word and Excel, let alone anything more safety or security critical. What happens when the train breaks down in a tunnel with no staff member on it (e.g. because the internet connection dropped)? Also, what happens when these staff go on strike? And a driver could earn extra by monitoring more than 1 train, if they wished. Sort-of possible when you get to the level of automation used on the DLR, but only when things are working perfectly and there are many situations where the attendant is required to operate the train from the front (including during high winds and when there are staff on or near the track). At all of these times they are required to give their full concentration to their single train. This is not something you can automate. Always a way of squaring the circle. Just because something can be done in theory doesn't mean it can be don in practice. And even when it can be done in practice, that doesn't mean it is economical to do it - e.g. would you rather spend billions to achieve the current service and safety levels or would you rather spend the same money on improvements? Old chestnut from the Airline industry; by 2030 [update year as required], the flight crew will comprise one man and one dog. The duty of the man is to Not touch anything. The duty of the dog is to bite the man if he fails to perform his duty :-) This saying has been around since (at least) the 1940s (the first autopilots were invented in the 1910s). That it still not come to pass, despite over a century of progress with aircraft automation, should make you realise that full automation is actually a much harder problem than it might seem from an armchair - and in many ways aircraft automation is a lot easier than railway automation (vastly fewer interfaces, much more controlled environments, and generally much greater margins of error). |
|
|
|
Post by andypurk on Oct 9, 2024 2:42:29 GMT
Just because automation is extremely difficult, doesn't mean it shouldn't start one day. Automation of the driving task on London Underground began in the 1960s on the Victoria line. Despite 60 years of investment and technological progress, drivers are still required. Why? Because running a safe railway with no staff onboard is an incredibly complicated task, made even more so by needing to interact with people in many locations. The obvious comparison is with airport people mover systems, but they are topographically extremely simple compared to even the simplest LU line (Waterloo & City), have lower usage and the track is completely segregated from passengers and other vehicles. A comparison with Paris is interesting, where a couple of the older lines have now undergone conversion to fully automated operation with no staff on-board. For the recent line 4 conversion this cost the best part of €0.5 billion for the 14 km line. However there are several features which have helped with the switchover of lines 1 and 4 in Paris including: Line 14 being automated from the beginning (giving experience), generally straight platforms for easier addition of platform edge doors, double track tunnels aiding in evacuation, availability of rolling stock which can be moved between lines as automation progresses. Closer to the UK, the Lille Metro has been fully automated since the beginning and the VAL system is in use in many other locations in France and elsewhere but these have the advantage of being new systems designed for unattended operation from the beginning. Introducing unattended operation on the Underground would need adaptation of the infrastructure for easier evacuation, which given the age of some sections, may need a lot of work for little benefit. Automation on the other hand is much easier with the majority of the Central London network already having automatic 'driving' but with the station operations under human control. |
|
|
|
Post by sealingmark on Oct 9, 2024 12:39:48 GMT
Thanks for your detailed, considered response. To be clear; my goal is not to get rid of drivers!
I see some automated train operation as a logical evolution. Go. Stop (in exactly the right place). Open doors.
Other tasks a driver currently performs may not be automatable. But could they be performed remotely?
Remote oversight and operation, either by computer or human, first requires a secure, robust means of communicating with every train. Surely not unachievable, as we are currently working very hard to provide mobile phone coverage to trains in tunnels.
My thought is that much automated operation could proceed unsupervised, but a human could be assigned as and when required. I think it's quite rare to need an evacuation from a train mid-tunnel; perhaps specialists could be dispatched, should that be required.
Perhaps ~90% (even 99%) automation could be feasibly achieved, whilst accepting that 100% is unfeasible.
|
|
|
|
Post by Dstock7080 on Oct 10, 2024 11:44:20 GMT
|
|
|
|
Post by Dstock7080 on Oct 17, 2024 15:20:37 GMT
RMT action now announced: www.rmt.org.uk/news/london-underground-workers-to-strike-over-pay-offer/1-2 November - Transplant/Engineering train staff 3-4 November - Track Access Control, Power Control and various other London Underground Control Centre grades 4 November - Emergency Response Unit 5 & 7 November - Stations and Trains grades 6 & 8 November - Service Control grades |
|
|
|
Post by spsmiler on Oct 20, 2024 11:44:11 GMT
sigh
|
|
|
|
Post by starlight73 on Oct 21, 2024 18:06:28 GMT
I’m assuming part of the reason for the management grades overtime ban is to prevent managers driving trains on the strike days. However, will it cause disruption on 9-11 Nov and 13-16 Nov? (i.e. days when there are no other RMT or ASLEF strikes) TfL’s website is saying: linkThat implies that the management overtime ban will not be noticeable by itself to passengers, otherwise their website would say ‘until 16 November’? Separately, I’m also wondering if the strikes by engineering, transplant (is this part of engineering?l, and the Emergency Response Unit (ERU) will affect passengers. I would’ve thought that an engineering strike simply means that the engineering work doesn’t get done and it has to be pushed back? if the ERU are on strike, does that mean services are likely to be suspended automatically? Or can trains run until something goes wrong, and then the problem will not get sorted? apologies if these are impossible questions as I might just have to wait for the TfL announcement |
|
|
|
Post by brigham on Oct 21, 2024 18:33:23 GMT
Driverless trains are clearly not the answer.
Perhaps we should try passengerless trains?
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Oct 21, 2024 22:14:19 GMT
I’m assuming part of the reason for the management grades overtime ban is to prevent managers driving trains on the strike days. Managers don't drive trains. Some, and I stress this is a small percentage of operational mangers, have an "emergency licence". This does not permit them to operate passenger trains in passenger service. On paper it's to allow them to move an empty train when no properly licenced driver is available but in practice I've never known a manager with such a licence to actually use it in anger in my 20 years as a driver. The idea of the overtime ban is to make life difficult internally. Say there's a gap in managerial coverage at the drivers book on point (where a manager must be present to book each driver on) - rather than having a manager cover it on overtime they'll have to use a manager that would have been doing background administrative work, carrying local investigations, disciplinary interviews, performance and development reviews, etc, etc with all that background work having to be caught up on at a later date. Separately, I’m also wondering if the strikes by engineering, transplant (is this part of engineering?l, and the Emergency Response Unit (ERU) will affect passengers. I would’ve thought that an engineering strike simply means that the engineering work doesn’t get done and it has to be pushed back? Strike action isn't always about inconveniencing the travelling general public. That having been said, if a new date has to be sought for strike affected works that will of course affect the travelling public on the new date - I guess you could argue its a form of delayed strike action. if the ERU are on strike, does that mean services are likely to be suspended automatically? Availability of the ERU is not critical to running a train service. London Fire Brigade (or County equivalents) are more than capable of dealing with emergencies on the railway. Obviously the specialist nature of the ERU is useful to have, but its not crucial. |
|
|
|
Post by brigham on Oct 22, 2024 7:38:51 GMT
"Managers don't drive trains."
What staff grade are the people who drive the limited number of trains running during strikes?
The ones which the Press describe as 'Managers'?
|
|
class411
Operations: Normal
Posts: 2,743 
|
Post by class411 on Oct 22, 2024 8:58:18 GMT
FWIW, I am not attacking train drivers, I'm criticising the system of industrial relations which results in the pain being inflicted on the travelling public, who aren't responsible and can't do anything about it. Corse the travelling pulic are responsible, they are the ones who elected government, both national and local. No, it's the fault of the cyclists. It's always the fault of the cyclists. |
|
|
|
Post by starlight73 on Oct 22, 2024 11:06:09 GMT
What staff grade are the people who drive the limited number of trains running during strikes? I think on National Rail trains there are managers with driver or conductor licenses who can work during strikes, but Colin says this isn’t applicable to LU. The limited LU trains which have run on strike days in the past are probably being driven by staff who aren’t on strike. (E.g. there is a day where RMT drivers are striking, but ASLEF ones aren’t, and vice versa). Presumably on the day when both RMT and ASLEF drivers are striking there will be no LU service. Thank you Colin! |
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Oct 22, 2024 11:41:18 GMT
There may be some drivers who are not a member of a union, and striking is not compulsory even for those that are members of a union that is on strike.
It is also theoretically possible for a former driver who has recently become a manager to still be current on their route knowledge, etc. but I can't imagine this ever being more than a single digit number of people at any given time.
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Oct 22, 2024 19:39:33 GMT
"Managers don't drive trains." What staff grade are the people who drive the limited number of trains running during strikes? The ones which the Press describe as 'Managers'? The press are hardly renowned for being factually correct and this is one of those situations. As answered by others, it will be non striking drivers that are driving the trains - either non union members or members of a union that hasn't called a strike. It is also theoretically possible for a former driver who has recently become a manager to still be current on their route knowledge, etc. but I can't imagine this ever being more than a single digit number of people at any given time. Again I've never known this to happen in all my years on LU. Once a driver is promoted to a management role, that's it; They're no longer a driver. If such a person was to drive a train on a strike day, I suspect the unions would have a view on it! |
|
|
|
Post by xtmw on Nov 1, 2024 13:13:50 GMT
RMT strike called off
No news on other unions
|
|
|
|
Post by Dstock7080 on Nov 1, 2024 13:41:02 GMT
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Nov 1, 2024 18:17:20 GMT
ASLEF are pressing ahead as planned. They will not suspend their action for further “talks” as they say there’s already been plenty of time for that (the current pay & conditions claim dates back to April this year). Only a positive change of offer from LU will get them interested in resolving the current dispute. |
|
|
|
Post by Dstock7080 on Nov 5, 2024 15:59:07 GMT
|
|
|
|
Post by Chris L on Nov 5, 2024 19:58:37 GMT
Thanks for your detailed, considered response. To be clear; my goal is not to get rid of drivers!
I see some automated train operation as a logical evolution. Go. Stop (in exactly the right place). Open doors.
Other tasks a driver currently performs may not be automatable. But could they be performed remotely?
Remote oversight and operation, either by computer or human, first requires a secure, robust means of communicating with every train. Surely not unachievable, as we are currently working very hard to provide mobile phone coverage to trains in tunnels.
My thought is that much automated operation could proceed unsupervised, but a human could be assigned as and when required. I think it's quite rare to need an evacuation from a train mid-tunnel; perhaps specialists could be dispatched, should that be required.
Perhaps ~90% (even 99%) automation could be feasibly achieved, whilst accepting that 100% is unfeasible.
Evacuations in existing tunnels would be extremely difficult without a safe walkway and staff supervising the process. You ignore the possibility of a track obstruction. Experiments were tried with cameras on the front of a train to spot such problems and stop the train. They were unsuccessful. Someone in the cab is needed. The initial Victoria line system was very much what you suggest but the trains didn't always stop in exactly the right place. |
|
|
|
Post by jimbo on Nov 5, 2024 20:52:45 GMT
Most LU lines now operate automatically, but with a member of staff for certain duties. Perhaps most of those tasks could also be automated, but what then does that member of staff do between rare emergencies when their role becomes crucial? With no regular physical activities, they are liable to day-dream and miss essential problems they are there to spot. The job will be unattractive to many if they turn up day after day to just ride trains without being called upon to use their expertise. The DLR train attendants can wander the train and interact with passengers, but that is rarely possible on a packed tube train. Without the saving of wages of an unstaffed train, the cost of full automation and its maintenance is hard to justify. It is said that the tube would require level platforms, gap fillers and platform edge doors. If we need someone in the front cab at all times, they need something to do all day!
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Nov 5, 2024 21:34:53 GMT
The reality is that the job of a train driver is extremely boring and repetitive, more so on an ATO railway. The ability to maintain concentration is therefore a key attribute. If one can't stay alert enough, they have no place on the railway!
As for being put off by an unattractive job role.......as long as it pays the money, the queue will continue to outstrip the vacancies.
|
|
|
|
Post by jimbo on Nov 6, 2024 1:27:35 GMT
The DLR train attendant method has been amazingly successful there. Introduced on opening in 1987 since there were 11 trains but 15 stations, it remains 37 years later despite the huge increase in both and deep level sections.
|
|
roythebus
Pleased to say the restoration of BEA coach MLL738 is as complete as it can be, now restoring MLL721
Posts: 1,275
|
Post by roythebus on Nov 7, 2024 21:56:27 GMT
The problem with dispatching ERUs, you have 3 trains stuck between two stations, each train packed to capacity. How does an ERU get to the middle train? It just ain't possible.
|
|
|
|
Post by jimbo on Nov 7, 2024 23:34:17 GMT
Would unstaffed empty stock workings be OK? Eg to/from siding or depot? If they get stuck, there is no worry about trapped public.
|
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Nov 8, 2024 0:51:53 GMT
I'm sure the unions would have something to say about it! Other than that, it will depend in part on how reliable the trains and automation is and how likely it is for other people to be in the vicinity of the train.
If it is likely a person on board is going to need to intervene in any way to get the train from A to B then they there wont be unstaffed movements. How disruptive a stalled train in the specific location will also factor into the decision. e.g. Blocking Neasden Depot throat would be a lot more disruptive than blocking access to Uxbridge sidings, so unstaffed operation is probably more likely at the latter than the former.
The greater the likelihood of people being around the train when it is moving the lower the likelihood of unstaffed operation, for obvious safety reasons. Depots are far more likely to have people nearby than sidings are, so again expect to see it in sidings before depots. If you ever see it.
|
|
