Driver-less trains time scale... (Metro)

[Bletchleyite]

If nothing else, even on AB lines the problem of making an automated rail vehicle is much simpler than an automated car. An automated rail vehicle has the simple rule - if in doubt stop - it doesn't have to consider swerving, nor pedestrians walking all over the place, nor cyclists etc, and the image analysis to identify if the track is obstructed is much, much simpler than identifying a path on a road. There are level crossings, but technology will develop in keeping them safe.

"Guard-only"[1] operation could well be the second saviour of the branch lines in my view.

A good place to test it would be Stourbridge, FWIW - a simple, small branch line with little or no interaction with other things. You could convert a 2-car D78 as a trial.

[1] Guard in this case = DLR style train captain with emergency <25mph on sight driving skills.

 

[Western Lord]

I agree the first few years will see slow up take, but I think it will not be to long before the growth becomes very rapid.

Cars are used incredibly inefficiently. Most of the time most, no almost all of them are sat parked and going nowhere. If you want to test this yourself, count 50 cars from your door at say 3 am, when the highest number of people are at home and then do the same at 8:45. The distance you have to walk will be a little bit further than in the middle of the night, but not much. This means that a huge amount of capital and resources are tied up doing nothing. The ‘ride hailing’ model allows vehicles to be pooled and used efficiently compared to the individually owned model, spreading the cost of the capital over many users.

Private cars do not have to make a profit. Most people regard them as an extension of their personal space and treat them accordingly. You don't have to take everything you need for the journey with you every time you get in the car and you don't have to remove all of your stuff every time you get out. The 'ride hailing' model is already available, they are called taxis. My car spends most of its time doing nothing, but it is available for whatever journey I want to do instantly 24/7. I am happy to pay for the freedom and convenience that this offers. I also choose what size of car and level of luxury I want. That will be virtually impossible with the pool car scenario, nobody could provide, almost instantly, different types of vehicle on demand.

 

[Bletchleyite]

That will be virtually impossible with the pool car scenario, nobody could provide, almost instantly, different types of vehicle on demand.

They will if ridership is high. Uber already offers several categories of vehicle.

I do agree with you on the car ownership thing though - it makes no economic sense whatsoever for me to own a car, but I have made a lifestyle choice to do so at a considerable cost.

But then it doesn't make economic sense for me to live in a 3 bedroom house either as there's only one of me. (Though I do have a friend as a lodger at the moment I don't know how long-term it will be).

 

[bramling]

If nothing else, even on AB lines the problem of making an automated rail vehicle is much simpler than an automated car. An automated rail vehicle has the simple rule - if in doubt stop - it doesn't have to consider swerving, nor pedestrians walking all over the place, nor cyclists etc, and the image analysis to identify if the track is obstructed is much, much simpler than identifying a path on a road. There are level crossings, but technology will develop in keeping them safe.

"Guard-only"[1] operation could well be the second saviour of the branch lines in my view.

A good place to test it would be Stourbridge, FWIW - a simple, small branch line with little or no interaction with other things. You could convert a 2-car D78 as a trial.

[1] Guard in this case = DLR style train captain with emergency <25mph on sight driving skills.

Stopping isn't always a consequence-free action. An emergency stop in poor adhesion conditions is likely to cause wheel flats, including on WSP-fitted trains. This costs money to fix, and could result in the train being unavailable for service. So one certainly doesn't want trains stopping all over the place just on the whim of a computer not liking something.

It all comes back to not whether driverless / ATO trains can be done, but whether the result is better and sufficiently worthwhile over what we have today. I'm simply not convinced of that.

 

[Chris M]

I live in inner, London, and from anecdotal observation around here the time when most cars are in use is actually weekends. During the week I theorise that people get public transport to work (the DLR is walking distance, and there are five bus routes within 15 minutes walk, and many more are less than half an hours walk away). Weekends people make journeys to many different places that are trickier by public transport (especially with engineering works) including to destinations like supermarkets and DIY stores where heavy and/or bulky loads that are less practical by public transport are quite common.

 

[Dave1987]

If nothing else, even on AB lines the problem of making an automated rail vehicle is much simpler than an automated car. An automated rail vehicle has the simple rule - if in doubt stop - it doesn't have to consider swerving, nor pedestrians walking all over the place, nor cyclists etc, and the image analysis to identify if the track is obstructed is much, much simpler than identifying a path on a road. There are level crossings, but technology will develop in keeping them safe.

"Guard-only"[1] operation could well be the second saviour of the branch lines in my view.

A good place to test it would be Stourbridge, FWIW - a simple, small branch line with little or no interaction with other things. You could convert a 2-car D78 as a trial.

[1] Guard in this case = DLR style train captain with emergency <25mph on sight driving skills.

Believe what you want to believe but I will maintain that you don't really understand what you talking about which is why I give you arguments no credence.

 

[notverydeep]

Private cars do not have to make a profit. Most people regard them as an extension of their personal space and treat them accordingly. You don't have to take everything you need for the journey with you every time you get in the car and you don't have to remove all of your stuff every time you get out. The 'ride hailing' model is already available, they are called taxis. My car spends most of its time doing nothing, but it is available for whatever journey I want to do instantly 24/7. I am happy to pay for the freedom and convenience that this offers. I also choose what size of car and level of luxury I want. That will be virtually impossible with the pool car scenario, nobody could provide, almost instantly, different types of vehicle on demand.

Taxis currently have a tiny market share of all journeys and almost no share of longer distance journeys - because with the driver's wages, they are hugely more expensive per marginal mile* than your own car. But if say 10% of the current vehicle fleet were available on a hail and ride basis, even in small towns, the response times (between request and pick up) would be sub 5 minutes – they really would be on every street corner (I accept that this won’t be the case in a hamlet 10km from the next village).

All this means that another of the key areas of improved efficiency of ARVs can be taken advantage of – I have an estate car, because I have sized it for the journeys that I do requiring the most space for luggage and Children, that is self-catering holidays once or twice a year. But mostly, my trips are with myself or just me plus one child with no luggage. With comprehensive ride hailing service, I could request and pay for the vehicle that is correctly sized for the trip. That might be Smart car size for when it is just me on a short journey (though the operator might still send a larger vehicle), but I would request more spacious and luxurious vehicle for the longer journeys. I would have this flexibility and would not have to pay £400 a month for the big luxury car I only get value from on the occasional long trip.

Despite the need for an operator profit, these efficiencies would mean that the cost per mile would be lower than the average cost per mile of a car you own and on a par with the marginal cost per mile - and that is usually less, often much less than a rail fare...

*The marginal cost per mile is not the same thing as the average cost per mile. This is because Purchase cost less residual value, Vehicle Excise Duty, MOT test and insurance are 'standing costs'. You have to pay these costs to do the first mile - you won't get any back if you only do one mile each year. It may well be that that first mile costs £700 each year, as my 8 year old Focus has for the last few years, but for me each subsequent mile can be had (and I have a spreadsheet to keep track) for around 17p.

 

[jon0844]

Private cars do not have to make a profit. Most people regard them as an extension of their personal space and treat them accordingly. You don't have to take everything you need for the journey with you every time you get in the car and you don't have to remove all of your stuff every time you get out. The 'ride hailing' model is already available, they are called taxis. My car spends most of its time doing nothing, but it is available for whatever journey I want to do instantly 24/7. I am happy to pay for the freedom and convenience that this offers. I also choose what size of car and level of luxury I want. That will be virtually impossible with the pool car scenario, nobody could provide, almost instantly, different types of vehicle on demand.

The belief that we'll all call a car and everything will be fine is a bit pie in the sky.

For one, it means cars driving around empty to pick up people - much like a taxi may - which doesn't seem a good use of resources, any more than the 'a lot of the time a car is sitting around doing nothing'. At the end of the day, you're maybe solving a parking problem but little more.

And as has been said, what about a car that already has your stuff in it? Coats, shoes, blankets, a car set and other bits and bobs? Are you going to take that in and out of every car you summon? What happens when the car you call is delayed, or turns up soiled? You go into your app and reject it, and wait for another?

It sounds good for some purposes, but I hardly see everyone giving up their own personal car - even if drives itself or not.

I am quite happy to buy a new car that has adaptive cruise control and can hold a lane, so I can be more hands-off when driving. I borrowed a car with adaptive cruise control and loved it, but it didn't steer. Frankly if it did both, I'd have run the risk of switching off and not being ready to step in should an emergency occur, so I think that's another problem - does a driver need to be alert at all times even if not driving? If so, good luck with that. The temptation to stop concentrating or looking ahead will be too great.

 

[Edinburgh2000]

The belief that we'll all call a car and everything will be fine is a bit pie in the sky.

...... At the end of the day, you're maybe solving a parking problem but little more.

And as has been said, what about a car that already has your stuff in it?

The market will demonstrate what people want. My belief is that there will be service providers offering 'packages', rather like you can get for your mobile phone or broadband, and you will choose what service you want. If you want a dedicated car, reserved for your use only, where you can leave your stuff, then there will be a price for that package. The packages are likely to include parking, and you will choose how far away from your home or workplace the car will park itself as part of your package (and hence the recall time for when you want to use the car).

A pensioner living in a semi-rural area might be very happy with a budget package that offers a car within 30 minutes, outside peak hours. Many would want a prompter service, and provision in the peak hours, and that will cost more. It will be up to the service providers to manage and plan the availability of vehicles so that the service levels are met for all its customers at the least cost.

But the 'my car only' package, with it stored in a parking lot most of the time, will be the most expensive package and many will choose to spend their money on something else.

 

[Bletchleyite]

Stopping isn't always a consequence-free action. An emergency stop in poor adhesion conditions is likely to cause wheel flats, including on WSP-fitted trains. This costs money to fix, and could result in the train being unavailable for service. So one certainly doesn't want trains stopping all over the place just on the whim of a computer not liking something.

It all comes back to not whether driverless / ATO trains can be done, but whether the result is better and sufficiently worthwhile over what we have today. I'm simply not convinced of that.

It's not about "better", it's about "cheaper".

Guards are cheaper than drivers. And one member of staff is cheaper than two, particularly if you can convince passengers it's OK because that member of staff will be dedicated to them rather than driving the train.

 

[Bletchleyite]

Believe what you want to believe but I will maintain that you don't really understand what you talking about which is why I give you arguments no credence.

To be honest, most of your arguments in anything relating to automation of anything (even TVMs, e-ticketing etc) are basically Luddism, so I doubt we are ever going to agree on these matters.

 

[jon0844]

It's not about "better", it's about "cheaper".

Guards are cheaper than drivers. And one member of staff is cheaper than two, particularly if you can convince passengers it's OK because that member of staff will be dedicated to them rather than driving the train.

Any improvements in technology that falls short of entirely driverless operation, but removes the need for certain work - and training (maybe less need to learn a route) will have an impact. If you can get driver training down to under a year, or down to six months or less, and reduce the cost of training - then you begin to have more power to fight the unions.

It won't be an overnight switch, and I believe that current drivers have little to worry about. It will be the next generation, possibly taken on with different contracts, that will be affected. Of course, depending on how the unions take to the idea!

 

[Bletchleyite]

Any improvements in technology that falls short of entirely driverless operation, but removes the need for certain work - and training (maybe less need to learn a route) will have an impact. If you can get driver training down to under a year, or down to six months or less, and reduce the cost of training - then you begin to have more power to fight the unions.

It won't be an overnight switch, and I believe that current drivers have little to worry about. It will be the next generation, possibly taken on with different contracts, that will be affected. Of course, depending on how the unions take to the idea!

Indeed. Like DOO on Merseyrail it'll come in slowly, so there's plenty of time for natural attrition as people leave and retire. So we might see traditional driver recruitment stop. But the future of the grade is by no means secure in my view.

 

[reddragon]

Fully autonomous cars will be legal on UK roads from next year, once the current bill passes through parliament. The real world cars will probably start from 2019, replacing taxis, rural bus services & private cars in cities first. Train services are not immune from this change by any means!

Car ownership is expected to drop 80% over the next 10 years as we book a car for a trip, either alone or shared. Many in road drivers jobs will not have a long future career!

Railways are different. It is much easier to make trains fully automatic, but also far more costly to make the railway suitable for autonomy, level crossings & passengers etc.

I would become a train driver knowing I could have a 50 year career, but knowing that full on driving will become auto pilot assisted, then monitoring and then a passenger interaction / safety role.

 

[Emblematic]

The belief that we'll all call a car and everything will be fine is a bit pie in the sky.

For one, it means cars driving around empty to pick up people - much like a taxi may - which doesn't seem a good use of resources, any more than the 'a lot of the time a car is sitting around doing nothing'. At the end of the day, you're maybe solving a parking problem but little more.

And as has been said, what about a car that already has your stuff in it? Coats, shoes, blankets, a car set and other bits and bobs? Are you going to take that in and out of every car you summon? What happens when the car you call is delayed, or turns up soiled? You go into your app and reject it, and wait for another?

It sounds good for some purposes, but I hardly see everyone giving up their own personal car - even if drives itself or not.

I am quite happy to buy a new car that has adaptive cruise control and can hold a lane, so I can be more hands-off when driving. I borrowed a car with adaptive cruise control and loved it, but it didn't steer. Frankly if it did both, I'd have run the risk of switching off and not being ready to step in should an emergency occur, so I think that's another problem - does a driver need to be alert at all times even if not driving? If so, good luck with that. The temptation to stop concentrating or looking ahead will be too great.

It's clear that private car ownership will continue for a long time - it's so convenient. It's also true that UK car ownership is in decline. If you look at the size of the car fleet against population, it becomes clear that per-capita ownership has gone into decline:

Year	Car Parc (m)	Population (m)
1996		22.2		58.2
2006		27.6		60.8
2016		30.9		65.6

That partially explains the growth in rail usage, but also shows how urban dwellers, who have reliable private hire and delivery services available, have started to give up on car ownership. Once automation drops out more of the costs, expect to see the trend accelerate.
Once automation reaches sufficient reliability, the human has no useful role in driving. The last thing you want the human doing is supervising the operation, they are really bad at it, their attention wanders if not sufficiently occupied. If anything, the right approach is the one where the automation supervises, with lane control and braking intervention. As I said before, the cost savings come when the professional driver is eliminated, this is where big business looks to recoup its investment. And to be honest, it wouldn't surprise me if automation eventually became mandatory, as the biggest safety problem on the road is the human. They kill thousands every year.

 

[158756]

Why though? Regional railways had its nadir too and its comeback is driven by very similar reasons. Regional train services have got faster and more frequent and connect smaller towns to larger urban areas which have services which can't be sustained in smaller urban areas. People in Hebden Bridge are still going to want to go into Manchester and Leeds city centres for the shops, and making vehicles autonomous isn't going to magically make them better at getting people there than the train. There really aren't many passenger rail lines left that don't ultimately fit a pattern of taking people on journeys which are somewhat difficult to do by car.

Services into the regional cities are much more at risk from driverless cars than London services are - congestion on the roads is nowhere near as bad as London so the trains have a much smaller journey time advantage, and a significant proportion of demand is made up of people who can't/can't afford/won't drive, rather than being forced onto the train by road congestion or London house prices, who would likely be amongst the first to jump if cars were cheaper/less stressful/didn't require a competent driver.

Hebden Bridge might be one of the better areas for the railway because the roads are so bad, but if driverless cars work out that way there would still be some modal shift, which would be a much bigger problem than in London because of the much lower passenger numbers to start with. Losing half the passengers on a pacer is a much bigger problem for the line's future than losing half the passengers on a class 700.

There's also the problem that if automation starts to put millions of office workers on the dole demand for both commuter and leisure travel will crash, and what's left will move towards the roads because of reduced congestion. Again London and intercity rail journeys would be most resilient because of the scale of congestion in London and the speed of Intercity travel.

 

[ModernRailways]

Looking for some advice.

I have recently passed my psychometrics and just waiting for confirmation of either being given a start date or being put in the holding pool to train as a driver for tyne and wear metro service.

My question is- What is the general opinion on how far away this type service is from being driver-less? or Light rail in the UK in general?

Im in a very stable career with prospects at the moment, its not quite driver money but its getting there. Becoming a driver is a dream for me but also a risk. i dont want to find in 10 years time computer automation being retro fitted into all stock and im out of a job! - iv got 35 years before i can retire!

Any advice would be appreciated or links to some relevant articles.
Thanks

I've just spotted this thread. Firstly, congratulations!

Metro won't be going driverless for at least 30+ years, the current plans for new trains have no driverless features. Most other light rail in the UK is trams, and as such they won't become driverless either, at least not until all the vehicles on the road are automated too. The current plans for Metro are being kept secret, but we've all been asked what we want to see from the trains, drivers have been asked what they want to see in the cab, and then passengers have been asked what they want the saloon to be like. From the rumours, a lot of things wanted are very similar to that of the S Stock in London, but Nexus have a weird desire for trams so who knows what we'll be seeing.

I think the closest we'll get to automation for a good 50 years or so, will be similar to that of CrossRail and the tube lines. Whereby the driver is in the cab and is still fully in charge of the train but the train drives itself for the most part. The DLR is probably the closest we'll get to full automation for a good 30 or so years unless we see a splurge in public transport funding with new systems popping up (which I doubt).

 

[bramling]

It's not about "better", it's about "cheaper".

Guards are cheaper than drivers. And one member of staff is cheaper than two, particularly if you can convince passengers it's OK because that member of staff will be dedicated to them rather than driving the train.

A marginal saving in the grand scheme of things. You're still employing a member of staff who has to ride with the train, so there are still all the recruitment costs plus it's still necessary to have the system of duty schedules (which is what today can so often make service recovery from disruption so difficult).

To balance this off, we have a massive technology cost, and the potential to end up with software companies being able to name their price. London Underground has found themselves bitten by this with the JNUP contract - any subsequent changes have cost a fortune when they have to go to Thales, and the imperfect system handed over but declared in compliance with the initial deliverables has resulted in specialist consultants having to be brought in, again one dreads to think of the cost of that.

I agree there is the potential to make a saving, but we're back to whether the benefits outweigh the disbenefits. Still not convinced!

 

[DerekC]

I think this may depend to a considerable extent on the way (train) drivers behave. The way the TOCs see it, anything which reduces the need for training and route familiarisation increases flexibility and potentially makes drivers more replaceable by others less skilled and lower paid. So the higher the drivers push the bar and the more difficult they appear to be, the more desirable automation of the driver role gets. It's not just a matter of straight economics - don't forget that politicians call the shots on what money is spent on railway research, and economics gets bent to suit what they want. And the problems of actually driving the train have already been solved, so it is perceived as easy to do. However several things will hold it back:

1. ATO requires an ATP underlay, and in the current world that means ETCS. ETCS is a good system, but it requires the signalling to be replaced and all the trains fitted on a route at (more or less) the same time - and every route has lots of modern signalling which doesn't need replacing yet. So the ETCS economics don't add up, so it isn't getting fitted, so ATO won't happen for a long time - unless someone comes up with ATO that works with the current signalling system, of course.
   
2. ATO requires an accurate data map of the network map which is kept up to date at all times. NR has been working on an integrated network model for about 15 years but so far without result.
   
3. The main line railway just isn't reliable enough for fully automatic operation - so a person on the train who can act as first line fixer of problems and can drive the train when needed will be needed for the foreseeable future.

 

[Olaf]

I think this may depend to a considerable extent on the way (train) drivers behave. The way the TOCs see it, anything which reduces the need for training and route familiarisation increases flexibility and potentially makes drivers more replaceable by others less skilled and lower paid. So the higher the drivers push the bar and the more difficult they appear to be, the more desirable automation of the driver role gets. It's not just a matter of straight economics - don't forget that politicians call the shots on what money is spent on railway research, and economics gets bent to suit what they want. And the problems of actually driving the train have already been solved, so it is perceived as easy to do. However several things will hold it back:

1. ATO requires an ATP underlay, and in the current world that means ETCS. ETCS is a good system, but it requires the signalling to be replaced and all the trains fitted on a route at (more or less) the same time - and every route has lots of modern signalling which doesn't need replacing yet. So the ETCS economics don't add up, so it isn't getting fitted, so ATO won't happen for a long time - unless someone comes up with ATO that works with the current signalling system, of course.
   
2. ATO requires an accurate data map of the network map which is kept up to date at all times. NR has been working on an integrated network model for about 15 years but so far without result.
   
3. The main line railway just isn't reliable enough for fully automatic operation - so a person on the train who can act as first line fixer of problems and can drive the train when needed will be needed for the foreseeable future.

There was a shift of stance about two years ago on the prioritisation on signalling enhancements - it was covered under the Digital Railway initiative - which will see the first fruits, if things progress as expected, announced by the end of this year.

 

[Dave1987]

To be honest, most of your arguments in anything relating to automation of anything (even TVMs, e-ticketing etc) are basically Luddism, so I doubt we are ever going to agree on these matters.

Ow I'm well aware that you think I'm a Luddite! Ticket machines are very unreliable, I see them out of order for various reasons all over the place. I also see it as one more wage packet lost from the economy, one more person who is on the dole or earning less money so either a burden on the state or contributing less to the economy and less tax but hey I'm a Luddite so what do I know.....

And with regards to driving of trains you clearly don't understand what you are talking about. I could rip you arguments to shreads but there is little point.

 

[Dave1987]

Any improvements in technology that falls short of entirely driverless operation, but removes the need for certain work - and training (maybe less need to learn a route) will have an impact. If you can get driver training down to under a year, or down to six months or less, and reduce the cost of training - then you begin to have more power to fight the unions.

It won't be an overnight switch, and I believe that current drivers have little to worry about. It will be the next generation, possibly taken on with different contracts, that will be affected. Of course, depending on how the unions take to the idea!

And you wonder why wages are stagnant in the economy.... I saw a economist talking on the Beeb news recently who was asked why wages simply aren't keeping up with inflation. He said jobs are so insecure now that people are willing to accept stagnant pay or even pay cuts to keep their job. Unions have been demonised so people feel intimidated for joining one.

 

[jon0844]

And you wonder why wages are stagnant in the economy.... I saw a economist talking on the Beeb news recently who was asked why wages simply aren't keeping up with inflation. He said jobs are so insecure now that people are willing to accept stagnant pay or even pay cuts to keep their job. Unions have been demonised so people feel intimidated for joining one.

Indeed. And drivers are very fortunate to have more power than most people could ever dream of, but that clearly can't last forever. Yet it will be a lot longer than most other professions, and it's absolutely true that people don't want to risk rocking the boat in their current jobs for fear of losing their job.

In journalism, about 11,000 people last year have given up. Many having been forced to go freelance, and work for ever lower rates. Workloads have increased and as people go, they're not replaced. Those who worked their way up the ladder are made redundant to be replaced with those starting at the bottom, who get paid loads less but are expected to do all the same work.

At my local Tesco, the supermarket has always claimed self checkouts have not made people redundant, but as more staff leave they are not being replaced. And at night, they turn off the scan and go machines so they don't need to have a member of staff on hand to do random checks and deal with other issues. Instead they use the self-scan machines and bring on a member of staff to man a till if required.

It's very depressing, but whatever we think, technology is still going to advance unless a Government steps in and says enough is enough and we need to give people jobs (and not just a universal wage to have us all sit at home getting fat on the sofa watching reality TV and getting ever dumber)...

 

[Billy A]

Just out of curiosity I looked up some figures for automation. Would the primary reason be to save on drivers' wages?

Apparently not.

The total cost of automating Line 1 on the Paris Métro was €600m. The headcount reduced from 240 to 30 if I recall correctly. If we assume €50,000 pa per head, a figure I've just plucked out of the air, the wage saving is about €10m pa. Not a decision you'd be taking just to save on wages then.

 

[ComUtoR]

Overtime, holidays, training, uniforms, pensions, sickness etc etc.

 

[Emblematic]

Just out of curiosity I looked up some figures for automation. Would the primary reason be to save on drivers' wages?

Apparently not.

The total cost of automating Line 1 on the Paris Métro was €600m. The headcount reduced from 240 to 30 if I recall correctly. If we assume €50,000 pa per head, a figure I've just plucked out of the air, the wage saving is about €10m pa. Not a decision you'd be taking just to save on wages then.

Not quite:

The Line 1 upgrade and full automation has cost €600 million. Of this, €400 million has been for the new trains, €150 million for the UTO and telecom systems, and the remainder for the installation of screen doors.

Source https://www.railengineer.uk/2012/02/28/paris-metro-line-1-a-new-beginning/

So realistically, taking the automation out may have saved up to €100m. Ten year payback.

 

[Billy A]

I had actually been wondering if the new rolling stock was included but when I read my source it seemed to suggest not.

It seems then that at least some train drivers need have some concerns!

 

[NotATrainspott]

Indeed. And drivers are very fortunate to have more power than most people could ever dream of, but that clearly can't last forever. Yet it will be a lot longer than most other professions, and it's absolutely true that people don't want to risk rocking the boat in their current jobs for fear of losing their job.

In journalism, about 11,000 people last year have given up. Many having been forced to go freelance, and work for ever lower rates. Workloads have increased and as people go, they're not replaced. Those who worked their way up the ladder are made redundant to be replaced with those starting at the bottom, who get paid loads less but are expected to do all the same work.

At my local Tesco, the supermarket has always claimed self checkouts have not made people redundant, but as more staff leave they are not being replaced. And at night, they turn off the scan and go machines so they don't need to have a member of staff on hand to do random checks and deal with other issues. Instead they use the self-scan machines and bring on a member of staff to man a till if required.

It's very depressing, but whatever we think, technology is still going to advance unless a Government steps in and says enough is enough and we need to give people jobs (and not just a universal wage to have us all sit at home getting fat on the sofa watching reality TV and getting ever dumber)...

The truth is that the days of mass employment being necessary for the economy to keep chugging along are nearing their end. That's why people's living conditions don't seem to match up with overall economic growth across the developed world. Ultimately, the economy could really just consist of the small proportion of the population who come up with truly novel ideas doing their work and letting the technologies they create provide for everyone else.

The UBI concept isn't about letting the rest of the population sit on the sofa all day though. It's about providing minimum standard of living regardless of what you're currently doing with your life. As a result, you can start to do things that you can't currently do so easily. For instance, people could head back into education, but not only to get things that they think will get them a job but to learn skills that they would actually enjoy having. Humans have a basically unlimited ability to invest time and money in utterly pointless activities and things which make them happier. A truck driver could start learning how to make pottery, find out that they're good at it, and end up running a daft wee business selling custom pottery once they're no longer needed in the cab. It's an utterly pointless economic activity, as a robotic factory will produce containers for things for a tiny fraction of the price, but not all humans want to have the cheapest possible thing at all times. If some people want to buy that pottery because it's well-made and attractive, then there will be a nice wee job making it.

It's best thought of as a permanent weekend. The sorts of activities which currently go on at the weekend will end up happening every day.

 

[DerekC]

There was a shift of stance about two years ago on the prioritisation on signalling enhancements - it was covered under the Digital Railway initiative - which will see the first fruits, if things progress as expected, announced by the end of this year.

Well, I hope you are right but there have been many false dawns before. From what I hear most of Digital Railway's money is now going into non-ETCS areas.

 

[Olaf]

Just out of curiosity I looked up some figures for automation. Would the primary reason be to save on drivers' wages?

Apparently not.

The total cost of automating Line 1 on the Paris Métro was €600m. The headcount reduced from 240 to 30 if I recall correctly. If we assume €50,000 pa per head, a figure I've just plucked out of the air, the wage saving is about €10m pa. Not a decision you'd be taking just to save on wages then.

Costs of employment in the UK are now over twice the basic wage due to legislation - those in France are even higher. Over the 35 year term of an investment amortization the benefits of automation and the capital investment out-weigh the risks when considering operational costs vs capitalisation.

The problem at present is in finding the initial investment funds.