The incident of the woman being dragged by the train shows that a driver actually being present didn't do anything. The woman got dragged which is the initial issue. The driver shouldn't have needed to intervene in the first place as the point of a driver observing the doors is to prevent this sort of thing happening in the first place. The obvious thing that should have happened is an emergency stop should have been pressed on the platform which stopped the train.
I don't see why driverless trains are a political gimmick when globally they are in use. If it simply was a political gimmick it would have never made it past the first city which introduced them. A lot of the points against driverless are effectively made null as the DLR manages fine driverless. What the public really need is a visible member of staff that moves up and down the train to help prevent problems before they occur. This can't be done from inside a cab doing nothing. Drivers themselves are a huge inefficiency, they are too slow to react to provide the intensive services that are required. The Victoria Line if it was manual would have been a total disaster as with the gaps some trains have between them there's no way a human would be able to stop and start the trains in time, only a computer really can. I'd honestly fear for safety if I knew a Victoria Line train was blasting through tunnels at 40mph with all those twists and turns and there was no way pinpoint way of knowing where the train in front was without waiting for it to appear in sight.
The other factor often used is having a one under, a sad reality is people do jump onto the tracks and can often fall. Going back to this a human again won't be able to stop the train fast than a computer. When you think about the fastest human reflexes, the ones that bypass the brain and are simply triggered by the central nervous system are 250ms at their best. Baring in mind pressing a stop button as you see a human on the track cannot happen through the central nervous system, such a reaction needs to go through the brain before potentials are sent down motor neurones which then activate the arm to press the button. A computer if triggered to do this by a sensor can at its slowest do it in around 50ms, with computers generally having reaction times of 10ms. So is a human in the front really worth it when a computer can be programmed to do the job far better? Such codes already exist and would need to simply be applied to trains. PEDs obviously being a better solution though, and should be fitted wherever possible.
I'm not advocating removing a member of staff from the train, however it's high time that the train operators got out of their cab and started serving customers instead of sitting in the cab doing something that a computer would do better anyway. When passengers need help on where to get off, when something goes wrong on board it's far easier to solve it when in the train as opposed to being secluded in a cab.
I also don't see what mass modifications would be needed at depots to adapt for trains that also have PEDs more than that needed for new stock on a normal day. Neasden depot serves 1996 stock and S Stock, of which the 1996 stock uses PEDs and the S Stock doesn't. While Ilford depot serves both 345s and 710s, of which the 345s use PSDs while the 710s don't.
Eastlondoner62 and @silenced I will seek you answer your points.
Firstly “political gimmick”. I do not mean that driverless trains are a political gimmick. They have an important place in the railway, particularly on new systems or existing systems that lend themselves to an ‘easier’ installation of driverless equipment. London Underground does not, Paris Metro did.
What I mean by “political gimmick” is that in the context of how politicians in particular have often thrown driverless trains into the mix specifically for London Underground, is that their comments in no way appreciate the challenges and risks that it would take to turn the whole network driverless. See the article from London Reconnections about driverless trains (not so heavy on the engineering aspect, risks and risk mitigation)…
www.londonreconnections.com/2021/the-political-myth-of-the-driverless-tube-train/Drivers and system
Firstly it must be remembered that drivers are part of the system. I think that is incredibly unfair to view well trained drivers as an inefficiency. Yes, all humans carry inefficiencies, and so do robots and computer systems, as at the end of the day systems/robots/computers are all designed by humans once you go full circle.
In regards to Clapham South, the Drivers input into the system worked. It did not fail. See the Rail Accident Investigation Branch report Report 04/2016: Serious accident at Clapham South tube station -https://www.gov.uk/raib-reports/serious-accident-at-clapham-south-tube-station . Note that no blame was attributed to the driver. In fact report highlights that the limitations of the in cab CCTV, limited what the driver could see (Quote from the report: The prompt actions of the train operator, using the in-cab CCTV system, may have reduced the consequences of the accident).
As I said earlier, the driver is part of the ‘system’. Therefore the driver can only be expected to be as good as the information that the system provides. In this case the information being fed to the driver from the in cab monitor and also the pilot lights authorising the drivers departure did not inform the driver that someone was trapped between the doors. In fact the drivers element of the ‘system’ succeeded as the driver followed the operational handbook to the letter, as the driver was pressing the go button continued to observe his in cab monitor as operational procedure requires until the last carriage has left platform. If the driver had not done this, he would not have picked up on the fact that something was wrong, and the woman would have continued to have been dragged, needless to say either resulting in more serious injury or loss of life.
The woman being dragged, was not the initial issue. When looking at a system you need to define where the system starts and ends. In this case, we can arguably say that the systems starts for the passenger at least once they enter network (station). What happened on the day that allowed for the platform to become so congested. Was it a lack of station control, and interruption in the train services, an incident that took place at a neighbouring station which passed on the issue to Clapham South etc. there are all sorts of possibilities.
Visible member of staff moving up and down a train
This mode of operation is OK on the DLR. Smaller trains, fewer people, shorter gaps between stations, predominately air sections and track layout that better accommodates detrainment. On LU, how would this work? I agree that customer service is an very important part of operating train services, however as I said in my previous response, is that the ultimate role of a driver is to attend to safety. Will a train captain/guard/attendant be able to do this role effectively whilst being in train carriages with more than potentially 500+ people? How will that impact on the safety of the member of staff and their ability to carry out key tasks such as liaison with service control, especially during service disruption?
Human reactions vs computers
Whilst your point
Eastlondoner62 makes sense in ideal conditions, many stations on LU are far from ideal. Important to remember that it’s not just those who are attempting to jump, it’s also people who are unaware of the dangers and stand too close to the track, people who suddenly take ill and faint, slips, trips, falls or even pushed.
In the scenario of straight wide platform with a clear yellow line that allows for computers to easily detect who isn’t to close to the platform, agreed that a computer would probably win over a human being. In the case of a cramped platform with a large curve (e.g. Bank), human intervention is often required. The computer would be far too sensitive and break at the slightest hint of passenger encroachment which happens all the time. Or it its sensitivity would be too low to be continuously effective. Hence favouritism for PEDS wherever go driverless. The LU incident log that is updated daily, provides numerous examples each month of quick driver reactions that have led to the preservation of life.
Stock modificationsAll depots when new stock is introduced on huge line upgrades which is what we are talking about when implement significant changes such as driverless technology need depot enabling works to be able to service new or modified rolling stock. By introducing PEDS you introduce new technology requires a different type of maintenance and testing.
Neasden Depot is effectively split into two. Stabling for Jubilee line 96 stock (servicing takes place at Stratford Market depot). S stock trains are serviced at Neasden along with stabling. When the S Stock was introduced the depot had to be reconfigured to enable handling of S stock. Same goes for other locations such as Hammersmith (Lilly Lane depot).
Waterloo & City lineCertainly, if you were going to start full driverless operation anywhere, it would make sense to start on this two station railway due to the simplicity.
Paris v LondonThis is mainly an engineering issue. Without going into this at depth I will use the analogy of London trying to become a cycling city like Amsterdam. Amsterdam was 'made for it', London was not. Hence why London is having such hard time trying to implement cycle infrastructure, and why it is so disruptive even after construction.
As one example, Paris is not a deep tube system generally with wider platforms and better ventilation being mainly cut and cover. In deep tube stations when you install new physical infrastructure the rules are more stringent around fire, ventilation and evacuation. Will or will you not use safeguarding such as PEDS? What type of PEDS will be used? How will PEDS change those dynamics in each station? What will be the scale of the mitigation? A complete platform rebuild like Bank or installation of more ventilation shafts to better cope with the spread of fire or a simpler update to the stations evacuation strategy?
Final pointsPolitical will. What I mean by this is that large upgrades cause huge and lengthy disruption especially where you need to physically remodel aspects of stations and platforms to make it happen. Are the public, government and authorities willing to put up with such long-term disruption? The amount of kick-back LU gets from attempting to close (whether fully or partial) a station to convert to step free access for 6 months alone is telling as to the level of NIMBYism that we have in this country.
@silenced, this has nothing to do with the French vs the English. It’s simply about the constraints our system has and whether it lends its hand or not to an efficient wholescale upgrade to driverless operation, whether CapEx costs are worth the end goal and who is willing to pay for it. It’s all good Boris Johnson preaching that driverless trains mean no more Tube strikes (which is untrue because signallers can go on strike and can be just as disruptive if not more) without actually understand the true scale of the cost to achieve it. This a lie, a "political gimmick". A major reason as to why the Paris metro suffers few to zero strikes, because they reformed their relationship with operations in 1990s.
I am not against fully driverless trains, far from it. Upgrade of train systems is the world in which I work, therefore I'm all in for new railway technology. However, I am against misreporting of the facts, overzealous meaningless statements from politicians and the lack of understanding as to how easily it can be achieved.
Apologies that this has gone off the topic of buses!