QueenslandRail Driver Only Operation (DOO) - How to DOO it?

[red dragin]

I know for a fact that Aurizon Redbank have made custom parts for the EMU's. Bet they aren't cheap.

[HappyTrainGuy]

Perth and Brisbane run the same models of trains.

I didn't know perth has EMUs, SMU200, IMU100, IMU120 and ICE trains. Should the TiltTrains also be factored into this aswell???

[#Metro]

https://en.wikipedia.org/wiki/Transperth_B-series_train

I seriously doubt that the high operational costs of QR can be explained by rolling stock. The simpler explanation is that with 2x the staff per train, there are 2x labour costs. Supported by the fact also that Sydney's costs are 2x that of Melbourne. Melbourne has DOO, Sydney does not.

Perth's trains made in the same QLD factory as QR's trains, and then transported to Perth.

A-series

Studies for the electrification of Perth's suburban rail network began in 1984 and in 1988 43 two-car electric railcars were ordered from Walkers Limited, Maryborough.

B-series

Manufacturer   Downer Rail/Bombardier
Built at   Maryborough
Entered service   2004

https://en.wikipedia.org/wiki/Transperth_B-series_train#/media/File:Transperth_Sets.JPG


https://en.wikipedia.org/wiki/Transperth_B-series_train#/media/File:B-Series_Interior.jpg

Is the Moreton Bay Rail line DOO enabled, or does that need ATP installed also?

[red dragin]

It was said somewhere there are about 300 drivers (so about 300 guards). Say they are on about $80k average, that's $27 million a year (Inc super).

Anyone know the annual SEQ QR budget is for train services and maintenance?

[verbatim9]

When are trains looking at becoming Driver only? Like Metro Trains in Vic?

[HappyTrainGuy]

Perth's trains made in the same QLD factory as QR's trains, and then transported to Perth.

A-series

Studies for the electrification of Perth's suburban rail network began in 1984 and in 1988 43 two-car electric railcars were ordered from Walkers Limited, Maryborough.

B-series

Manufacturer   Downer Rail/Bombardier
Built at   Maryborough
Entered service   2004

https://en.wikipedia.org/wiki/Transperth_B-series_train#/media/File:Transperth_Sets.JPG


https://en.wikipedia.org/wiki/Transperth_B-series_train#/media/File:B-Series_Interior.jpg

What does the factory have to do with it??? Just because they came out of the same factory doesn't mean they use the same parts or the parts are compatible across the entire fleet. As I said before. Perth doesn't have QR EMUs, QR SMU200s, QR IMU100/110s or the QR ICE sets despite them all coming out of the same factory located up north. The EMU and SMU200 share different door components for example - listen for the click/lock in an EMU once the doors close. The traction motors on a EMU01 are different to the ones on IMU105. Brake pads are different between SMU266 and SMU226. Air conditioning units are different across the fleet. The glass windows are different across the fleet. Seats are different. Electronics are vast.....vastly different across the fleet. Sure the SMU220/260 and IMU160 might be similar when compared to the transperth versions but the last time I heard Perth has 2 different depots just to service each particular generation of train. We just have 1 depot that services the tilttrains, EMU, SMU200, SMU220, SMU260/IMU160, IMU100, IMU110 and the ICE sets.

Not just rollingstock. And what about the size of the network? And the associated cost of having staff to sort out problems should they arise such as QR looking after the passenger and freight network in SEQ.

[#Metro]

Not just rollingstock. And what about the size of the network? And the associated cost of having staff to sort out problems should they arise such as QR looking after the passenger and freight network in SEQ.

The problem with your explanation is that it can't explain why the same relationship also exists between Melbourne and Sydney. The DOO theory can explain that.

[HappyTrainGuy]

Is that your assumption or fact?

[#Metro]

Is that your assumption or fact?

It is inductive reasoning.

Inductive reasoning (as opposed to deductive reasoning or abductive reasoning) is reasoning in which the premises are viewed as supplying strong evidence for the truth of the conclusion. While the conclusion of a deductive argument is certain, the truth of the conclusion of an inductive argument is probable, based upon the evidence given.

https://en.wikipedia.org/wiki/Inductive_reasoning

Further information could confirm or disprove the idea, but so far it explains a lot more than your explanation.

[HappyTrainGuy]

In other words its still your assumption all dressed up for a night on the town.

[#Metro]

In other words its still your assumption all dressed up for a night on the town.

Your explanation does not explain why the same relationship is seen between Melbourne and Sydney. Mine does.

And if you have 2x staff, one would not expect to hold 1x costs. You would expect to have 2x costs.

Why would one expect a doubling the staff levels to have no impact on overall production costs?

[HappyTrainGuy]

No it doesn't. Its your assumption based on something with so many factors which can easily influence the data. The same data that you are basing your dressed up assumption on. And going to a DOO system compared to the driver + guard does not equal double the cost which for some reason you seem to believe. On paper maybe but in real life and in a real railway its simply not the case.

I could just as well say that's because in Melbourne buses feed heavily into the railway stations combined with metro running the rollingstock and infrastructure into the ground which is creating all the issues with trains running express to meet otp whereas in Sydney they have a similar problem that Brisbane has where buses do not feed into railway stations combined with the aging rollingstock needing more attention.

If you want to use Brisbane as an example too we have a chronic lack of feeder buses, any actual buses, pt blackholes with an abundance of buses duplicating areas that don't even need them. That is affecting the numbers that use trains which in turn affects the cost ratio. And the public transport use among the population also impacts on this. High fares or lower fares. Lots of users or lack of users. Fast to use or slow to use.

Other factors like rollingstock costs play a part. For examples sake lets say it costs $1000 for an EMU to run 100km while the new SMU260 only costs $900.

There are just way too many factors to consider which can widely influence the data depending on how you want to look at and interrupt it.

[#Metro]

No it doesn't. Its your assumption based on something with so many factors which can easily influence the data. The same data that you are basing your dressed up assumption on.

I could just as well say that's because in Melbourne buses feed heavily into the railway stations combined with metro running the rollingstock and infrastructure into the ground which is creating all the issues with trains running express to meet otp whereas in Sydney they have a similar problem that Brisbane has where buses do not feed into railway stations combined with the aging rollingstock needing more attention.

But your explanation has even less evidence going for it.

And your bus feeding explanation is testable. Just look at how many people are on the rail network between Sydney and Melbourne.

This is simple to do.

Sydney has about 310 million trips in 2013, and 316 million in 2014 http://www.bts.nsw.gov.au/Graphs/Rail/Patronage

Melbourne has about 225.7 million trips in 2013-2014. https://www.ptv.vic.gov.au/about-ptv/ptv-data-and-reports/research-and-statistics/

Sydney has more people on the train network, therefore when dividing total costs per person, we should expect that it should have the lower cost per passenger trip made. 

It does not. It is 2x that of Melbourne.

Melbourne's patronage isn't 2x. It is more like 0.75x that of Sydney.

So we can rule out differences in network patronage as the cause here. It does not explain the (large) difference because both cities have broadly similar patronage on the rail system.

And we know that labour cost is a very large input cost:

The lack of a driver is the key to those extreme frequencies.  When you have a driver on every vehicle, the labor cost is the dominant cost of operations.

Frequency and Freedom on Driverless Rapid Transit, Human Transit
http://humantransit.org/2010/02/driverless-rapid-transit-why-it-matters.html

The problems we have seen with the Moreton Bay Rail line in both the operational cost (generally) and the training lead times suggest that BCC would be well advised to go automatic for its proposed metro.

Indeed part of me wonders if the MBRL could be partially automatic, crew step off at Petrie, send the train down the line in auto mode...

[petey3801]

DOO and ATO are two completely different beasts, with ATO requiring significantly more capital to get, especially on an established suburban railway.
Another question regarding thise figures, do they include cost of extra station staff, extra security (Perth have railway police on many stations during the evening/night, at least two on every Mandurah line station when I was there last and Melbourne is using PSAs quite heavily across the network due to a major increase in anti-social and criminal behaviour).
Also need to remember that the Perth network was completely rebuilt during the transition from diesel to electric (the system was on the brink of shut down), making the network quite fast (compared to Brisbane) and has all straight, high level platforms. This means drivers can do more runs in a shift, meaning less drivers are required for the carriage of more passengers.
Melbourne trains have the driver exit the cab to put on wheelchairs at the leading door - something that is a massive waste of time (in Brisbane, drivers must isolate and secure the cab before leaving it - a policy that may or may not change if DOO is ever implemented) and also won't work here due to many platforms being too narrow to do that at the far ends. Even the high level platforms still need a small ramp for wheelchair pax due to the horizontal gap.
Personally, I prefer to have guards left on the trains,but am open to having their role expanded with more customer service etc on board the trains. When s##t hits the fan, it is much more comforting to know there are two people looking after the safety of up to 1000 people on the train than one, especially if there is an emergency evacuation required or there is a fatality/other major incident. Last thing I want to be doing after a major accident/fatality is to be making PA announcements, dealing with agitated passengers (mmajor delays etc, and yes, some people to get very angry at staff when there is a fatality).
As long as the EMU, ICE and SMU200/IMU100 are here, the guards will be on them. These trains do not 'talk' to each other with regards to faults apart from an indication telling the driver that there is a fault. The fault needs to be found and rectified from the unit that it is on. With EMUs in particular, manual fault rectification on the move is often required with the guard operating switches etc in the rear unit under the drivers direction while still keeping the train moving. No guard - train is cancelled, cause it is simply not feasible for the driver to isolate the cab and walk to the far end of the train at every station to continue fault rectification.
EMU/ICE should be gone in a few years, so that is one hurdle gone in the shorter term future, but 100/200s future is still up in the air.

Can we make a list of pros and cons for DOO?
Pro:
- Reduced operating cost (including reduced cost of adding services);

Con:
- Less onboard customer service;
- Problems with wheelchair loading/unloading (extra station staff required? $$)
- Major Capex required to retrofit stations and trains for DOO (higher priorities for that $$ elsewhere?)
- Reduced ability to quickly deal with onboard problems (medical emergencies, problematic passenhers etc) and lost property (relatively minor issue)

Feel free to add to the list, although things like 'increase services' as a pro for DOO isn't really a pro IMO, as services can be increased with a guard on board too, just costs a bit more (covered under 'reduced operating costs'. Edit: Have added 'reduced cost of adding services' with 'Reduced operating costs'). Plenty of other heavy suburban railways in the world who have two staff onboard trains and still run high frequency trains! Just a matter of priorities.

[SurfRail]

This debate is completely pointless in the absence of information about QR's operating budget. 

Maybe instead of making suppositions, if people want to make arguments about line items in the budget they could maybe look into finding out the information from QR.

[#Metro]

Personally, I prefer to have guards left on the trains,but am open to having their role expanded with more customer service etc on board the trains. When s##t hits the fan, it is much more comforting to know there are two people looking after the safety of up to 1000 people on the train than one,

ATO is established technology. Automation also coming to cars and soon after that buses.

Lots happening with machine learning and artificial intelligence as well. Nobody's job is secure, especially dangerous, expensive or tedious ones.

https://vimeo.com/188105076

It is only a matter of time really.

Plenty of other heavy suburban railways in the world who have two staff onboard trains and still run high-frequency trains! Just a matter of priorities.

Under a fixed budget, twice the staff means half the service for the same input money. Transformation will take time, but that's not an impediment to its eventual installation.

Change is expensive - but most things that government does is expensive. That just goes with the territory of being Government. Costs are large, but so also tend to be the benefits.

Perfectly happy with Melbourne security guards during the night - QR also has security guards in addition to drivers, so this idea that if you have one you won't have the other doesn't hold water.

If you want security, get a security guard.

Information is much more readily available with mapping apps on phones, and the internet. This is the reason TransLink has modified their home page, for instance. So the role for a 'customer service factor' is declining IMHO.

I rode a train in Copenhagen, Denmark. There was nobody on board. You can sit in the front window while it is whizzing through the tunnels. Vancouver has this too.

It looked like this:

And I think even Gina Rinehart's trains and trucks are also automated.

[petey3801]

Nowhere did I say ATO is impossible, I said it is a very different beast to DOO and requires significantly more $$.
Also, we don't have a fixed budget, budgets are changing all the time, so not really a good arguing point.

Melbourne don't use security guards as such, they are basically railway police (authorised to detain, even carry weapons I believe). Brisbane security guards are basically powerless visual deterrants.

Some customer service roles are declining, but others are not.
Ginas railway is to be manned, trucks are possibly automated (not 100% sure). The ones you are likely thinking about is Rio who have automated trucks and are working on ATO for their trains (and very much still struggling at the moment).
I know the world already has ATO, have traveled on them myself plenty of times. BUT, these are mostly new build railways and these days, for ATO to be implemented, require basically a completely closed off system,unable to be accessed by trespassers. Doing that to the QR (or nasically any other suburban heavy rail network) will be a tremendous cost until the automation is to a degree that it can sense whether there is something on the tracks that should not be there, from a good size distance. Rio don't have to worry quite as much about human trespass as the Pilbara really is in the middle of BF nowhere, whereas the middle of suburbia faces many other problems with trespass and vandalism.

[#Metro]

UTIP has some interesting info:

For conventional lines that upgrade to UTO in parallel with the renewal of rolling stock or signalling equipment, it is estimated that the return on investment period is around 10 years. (The automatism costs being offset by gains in rolling stock fleet, this
figure refers to the extra cost of retrofitting PSD into existing stations). For more details on conversion, see
the dedicated section to Paris Line 1.

Operational cost factors: staff & energy gains

When factoring in operational costs, automated lines come clearly ahead of conventional lines; some studies
indicate a halving in operational costs. Staff costs are greatly reduced thanks to the abolition of the drivers'
function, even in cases of line conversion, when staff is likely to be retrained and deployed to other
functions. Acceleration and deceleration patterns can be adjusted to reduce energy consumption and
maximise energy recovery, thus significantly reducing energy costs. While maintenance costs are marginally
increased due to the introduction of platform and track protection systems, the overall balance is positive
thanks to the gains in personnel and energy costs.

http://www.uitp.org/sites/default/files/Metro%20automation%20-%20facts%20and%20figures.pdf

Apparently, Paris Line 1 was converted from manual operation to UTO.

Line 1 (as indicated by its name) was the first line to open, with its inaugural section opening in 1900. It is also the first line on the network to be converted from manually driven operation to fully automated operation.

https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_1

for ATO to be implemented, require basically a completely closed off system,unable to be accessed by trespassers.

It would require higher security to stop people getting into the tracks, however, people already get into the tracks now and trains are limited by the physics of their stopping distance already.

I noted that not all the stations on the Copenhagen Metro had platform screen doors. Some were lacking them.

I agree, Rome wasn't built in a day, but neither was the network we have today.

[#Metro]

No room for train guards in driver-only view of Sydney trains future
http://www.dailytelegraph.com.au/news/nsw/no-room-for-train-guards-in-driveronly-view-of-sydney-trains-future/news-story/4b357bc6feff1450679ce8b8fee98ad2

THE state-of-the-art trains set to replace Sydney's ageing fleet may operate with no staff other than its driver.

The New Intercity Fleet (NIF) will service passengers from Central to the Hunter, Central Coast, Illawarra and Blue Mountains with 65 ­double-decker trains.

Each of the 512 carriages will use cutting-edge transport technology that could include different seat configurations, wi-fi, bike storage, luggage racks and premium options.

But they may only have one staff member on board — the driver.

[#Metro]

Daniel Bowen on the removal of guards from the Melbourne train network in 1994-1995.

The Melbourne network is quite old, so it is interesting to reflect on how they managed to solve or overcome the DDA issues.

If you are putting more staff into PT, you wouldn't prioritise train guards
http://www.danielbowen.com/2013/02/04/train-guards/

Do we need them back?

Was the removal of guards a bad thing?

On reflection, I would say no.

There have been some incidents (though as far as I recall, no injuries or deaths), particularly with prams and small children, at risk from doors closing too quickly, particularly when not boarding (as recommended) near the front of the train.

[...]

But I think there is a strong argument to be made that guards were of little use to passengers — employing more drivers to cut waiting times and crowding by providing more train services makes a bigger difference.

So if you're going to put more money and staff into the train system, drivers and maintenance staff are far more useful to most passengers than guards.

[Gazza]

When factoring in operational costs, automated lines come clearly ahead of conventional lines; some studies
indicate a halving in operational costs

.

With the paris metro, they went from 1 driver to zero drivers, but the reduction in running costs was half.
That says the other 50% is energy, maintenance, control room staff, security etc.

That sort of backs up my guesstimate in the other thread thread that stated that removing guards might save around 1/3 of costs, because the other non labour related overheads still remain and are perhaps more difficult to cut costs in the short term.

[#Metro]

That sort of backs up my guesstimate in the other thread thread that stated that removing guards might save around 1/3 of costs, because the other non labour related overheads still remain and are perhaps more difficult to cut costs in the short term.

It is still a guesstimate, likey no more accurate than my 2x (I mean 1.93x).

Would need to go to a proper study and look at different options. For example, new lines first etc.

Stations could be upgraded faster if BCC allowed TOD zoning around stations and development around stations were used to fund station upgrades to DDA standards, straightening etc.

But I think we both agree it would achieve substantial cost savings, which could be later spent on service upgrades.

May increase driver tasks, but if ATP comes in and lineside signalling is removed (as the QLD Gov't media release suggests) then that extra work could be perhaps offset a bit. Would also like to see more automation of trains driving between stations.

[Gazza]

It would be more accurate than your 1.93 because 1.93 assigns all other train running costs a value of zero which a simple thought experiment proves cannot be true (Energy is not free, wear and tear costs money to repair, trains need someone in the control room etc)

[#Metro]

It would be more accurate than your 1.93 because 1.93 assigns all other train running costs a value of zero which a simple thought experiment proves cannot be true (Energy is not free, wear and tear costs money to repair, trains need someone in the control room etc)

Well, true, I should have just wrote "labour costs".

I note that you copy my lines. As the saying goes "Imitation is the sincerest form of flattery."

[BrizCommuter]

You can't compare automation of a self-contained, fully segregated metro line (e.g. Paris Line 1, Copenhagen Metro) with a non self-contained, non grade-segregated suburban rail system such as Brisbane. Completely different kettle of fish!

Please find some examples of ATO on suburban rail systems on anything other than an underground network core (e.g. Paris RER A). Note: you won't find any!

[#Metro]

I think STO should be investigated. It is not driverless and it is not fully automatic. Driver still has to stop the train, open doors etc.

Justifying automation
http://www.railway-technology.com/features/feature127703/

Automatic train operation (ATO) covers a range of levels of automation and can be widely differentiated in semi-automatic train operation (STO), driverless train operation (DTO) and unattended train operation (UTO).

With STO, implemented for instance on the London Victoria and Central line in the UK, the operation of the train's motors and brakes is automated, allowing a more consistent form of driving.

The driver however remains in the front of the train, to operate doors, give start signals when leaving a station and monitor performance of the train and the track ahead.

Unlike metro applications, for Main Line operation ATO is not "all or nothing". There
can be various in-between cases, not necessarily requiring precise stopping in
stations. Frequently, this provides "cruise control" allowing the driver to select a
constant speed ; "intelligent cruise control" that follows a speed profile taking into
account speed restrictions, gradients, timetable and energy saving (e.g. on the High
Speed Lines in Spain). These systems are aids to the driver (as in STO mode, with
several additions to it) but do not necessarily stop the train automatically at stations.

IRSE INTERNATIONAL TECHNICAL COMMITTEE
ATO for Suburban and Main Lines
http://www.irse.org/knowledge/publicdocuments/2010_04_30_ATO_for_Suburban_and_Main_Lines.pdf

[MichaelJ]

Our new North West Rail Link in Sydney will have no Driver as such on board but as stated earlier a sealed system is required for automation.  There's no ifs or buts about that one.

Driver Only Operation, could it be done in Queensland?  Yeah, it could be done but a two person Crew will always be superior.  Superior customer service and superior safety (for staff and passengers are the pivotal reasons for it.  Whether the 'second person' is a Train Guard (safeworking role) or a Security Guard or a Transport Officer - there are still significant labour costs involved.

[#Metro]

I would like to see a proper investigation into semi-automation or DOO on the QR network.

It may be that some lines go DOO (i.e. Gold Coast - Airport) progressively.

We should also raise platform heights. Odd to think that platforms are low because decades ago we had super-low trains, but they haven't

raised the platforms since.

[MichaelJ]

I agree, an investigation in to DOO would be a good thing for QR - so they can finally put the idea to bed and leave it.  In saying that, QR Guards need to pick up their act a bit and actively seek additional duties.  The difference between what a Sydney/NSW Trains Guard is required to do and what a QR Guard is required to do is astounding!

Platform heights - it's embarrassing that in this day and age, entire platforms weren't raised during station improvement works (thinking Petrie and Eagle Junction in particular).  They all need raising but I'd say they could get it done over the next decade or so to spread out the cost.

[tazzer9]

Platform heights - it's embarrassing that in this day and age, entire platforms weren't raised during station improvement works (thinking Petrie and Eagle Junction in particular).  They all need raising but I'd say they could get it done over the next decade or so to spread out the cost.

Please don't go out on the Ipswich line, you have have a heart attack when you see what they are doing with dinmore and graceville. 
I was just in shock when i last went past graceville and saw that only 20m of the platform has been raised, despite that massive amount of work, including repaving along the entire platform. 

EJ is a bit of an outlier as the massive curve likely prevents platform raising, petrie might be similar but the curves are less extreme.

[MichaelJ]

Platforms on curves can still be raised but it requires further engineering work such as moving the track to retain the kinetic envelope.

[SurfRail]

10 stations a year is hardly breakneck speed and would see the entire system finished in about a decade and a bit - especially if the stations beyond Caboolture and Ipswich are pushed to the back of the queue.

[#Metro]

How much does platform or a new station cost?

If we link the station changes to TOD development (like an actual TOD development push, not ad-hoc 'boutique' projects) I wonder if we could speed things up / have more funds for this.

TOD hasn't really been pushed much in SEQ IMHO. A good opportunity to raise platforms and do stations up I think.

[verbatim9]

^^Driver only can't come soon enough especially with a wage blow out, some drivers earning up to $100,000 per year" Is it necessary getting all platforms raised and compliant for Driver only operations?  In Mel not all platforms are level driver still needs to operate a manual ramp. Could the exsisting camera network on station platforms synchronise with the driver cab? Therefore minimal network of new cameras would need to be installed on platforms.

[#Metro]

Raising only half or part of a platform should be banned.

We cannot have half-a**ed NEW platforms going in all over the network just because decades ago we ran low trains.

Bring on TODs to pay for these upgrades.

[HappyTrainGuy]

Raising the platforms is a state funding issue. And their are many of them.

[SurfRail]

There are of course stations which should just be closed in lieu of spending any money on them whatsoever.  My list remains, at minimum:

- Holmview
- Gailes
- Wynnum
- Bindha

The money saved goes into upgrading the nearby stations (including expanding parking capacity as required).

[verbatim9]

There are of course stations which should just be closed in lieu of spending any money on them whatsoever.  My list remains, at minimum:

- Holmview

^^Yep remove this station and build a premium Beenleigh station and a new Bus Interchange similar to what's happening at Helensvale. And a wider cutting North of Beenleigh under the road to allow for 4 tracks Holmview to Beenleigh.

[MichaelJ]

We put passengers requiring boarding assistance in the sixth car so it's halfway between the middle and rear position for the Guard. I suggested the fifth car for QR but got told it wasn't really possible; 1) the existing humps in the middle and 2) the amount of infrastructure often at the fifth position car. The first shouldn't even have been a possible reason!

If we're going to look toward raising platforms in a more hastily manner, I think an effective mix of patronage and curvature needs to be considered.

Further, to speed up the station stop, station staff should provide the ramp to passengers requiring boarding assistance. I remember one day, I had seven wheelchairs going to the City and they delayed the service maybe 30 seconds because station staff were all over it. It was brilliant!

[#Metro]

Wacky Dude thoughts, but I write them down here nonetheless:

I am thinking about the "problem" with the NGR trains. And I am wondering if perhaps the NGR trains could be placed into service on the Gold Coast - Brisbane Airport line??

The Gold Coast-Brisbane Airport Line could then be run as DOO. Most stations are new(ish) and the ones that would need to be upgraded to DDA standards would be minimal this way.

Not sure how this would fit in with the ATP side of things either, but how much would ATP cost on the Gold Coast - Airport
Line to install??