Sydney Metro - Tallawong to Bankstown

[rtt_rules]

Operating costs for metro have been kept commercial in confidence so far, but we might get a crack at seeing them this year. Overseas experience is that unmanned automated commuter railways reduce opex by 30 to 50%.

Automated trams are being trialled in Europe and a couple of other places. Already they are parking themselves in depots and termini and soon will be in revenue service on the street, initially with an attendant on board and eventually driverless. The new Sydney tram system with its separated lanes is perfectly placed for full automation. Buses are in the same position as cars. Automation has already started. It's just a process of moving forward to confident and safe unattended operation. It will start to come this decade. The notion of a commuter train with two staff on board will be like something out of the Flintstones. Perhaps the Rail Transport Museum can take over the operation.

There's no disputing the fact that driverless automated railways reduce Opex on new lines or potential converted lines if feasible, and that's not the issue, but is it that simple when taking into consideration Capex in converting complex surface legacy rail networks? That includes the economic cost of disruption to existing rail services during conversion, which doesn't appear to be taken into consideration from what I can see? That's why I'd like to see a genuine cost/benefit analysis compared with GoA2 conversion with DOO and not based on some ideological whim.

Driverless cars, buses and trams, although already being trialled, haven't yet been proven to be capable of being operated safely in a real world street environment. Perhaps they might be one day, but that could still be a long way off. In the meantime.

Automated trams is interesting. I'm thinking for Melbourne type tram operations this would be a bit more challenging, however for Sydney, Brisbane, Canberra and Adelaide type operations, ie LR more than tram it would be fairly straight forward.

Do they de-man the tram completely or just move the staff member into the body of the tram checking revenue etc?

[rtt_rules]

Sydney trains will not be converted to Metro, however it is likely to walk slowly up the automation steps starting with ATO.

I can agree with you on that score, but unlikely to advance beyond GoA2 for the foreseeable future. It's already underway on T4 as the first stage.

I would think that the major issue with GoA4 (driverless) on a surface railway and operating alongside other rail uses is the need to safely isolate it from intrusion along the rail corridor, which isn't as easy as an underground or elevated railway. I don't know if the technology is yet sufficiently advanced to allow for fail/safe operation in that environment and the Bankstown Line conversion may be somewhat experimental and hence the complexity. Happy to be proven otherwise.

Infrastructure upgrades like platform levelling, screen doors and gap fillers really aren't an issue and from what I can see have been completed. The OH and power source are the same, although as Linto pointed out earlier, power upgrades and GoA4 resignalling may not have been completed, no doubt because of ETU industrial action. But they nevertheless still require significant shutdowns to existing rail services.

Even if GoA4 becomes proven technology for legacy rail networks in the future, it will come at a price. Rail operators may still prefer to go with GoA2 with drivers, which still gives a high level of service at less expense. New lines are a completely different matter.

Dubai Metro has a ground operation section and if I recall also Vancouver. I personally don't see this to be an issue, put up a fence and job done. I suspect less people play chicken with a train with no driver than with and if they want to and loose, the greater commuity won't shed a tear. However ironically we all know neither the computer or the driver can make the train stop faster.

I think the main concerns are during the phase in, there will be automated trains mixing with manual driven, but ATO basically makes the train auto once it leaves the station and I think the French did a phase in approach. But there is also country and freights, which by then (talking mid 2030's) will be less mixing with these as more dedicated freight lines are built. However technically I don't see an issue anyway. Both trains follow the same safe working system, what ever that is. AND who is to say that by then they are not also being automated?

The steps I mentioned will take decades, ATO needs to be implemented first and how many years before this project is complete? I suspect once this project is fully complete, then the next step will be placed on offer.

[rtt_rules]

The disruption to Sydney's other rail services for the closure is minimal and no worse than any other MTCE and upgrade work.

Compexity appears to be driven by union issues, less technical. Certainly working on up to 100 year old infrastructure has its risks to budget and schedules but if we always said its all too hard even ATO wouldn't proceed as some of the work required for Metro is also the same for ATO.

Primarly : The line is being removed from ST to free up slots for T8/T2/T3 in the city.
Secondary : Conversion to Metro as part of modernisation of the line and because M1 needs to run somewhere amd M1 is being run to Bankstown to enable densification of the corridor to cater for population growth of Sydney.

You clearly have no idea. The disruption is caused on the line that is closed and has got nothing to do with other lines.

It's convenient to blame industrial action for complexity and delays, but I don't buy it. There's more to it than that. As for 100 year old infrastructure causing problems, that's also misleading. About the only thing being 100 years old is the stations, most with curved platforms, but they have already been upgraded with platform levelling and installation of platform screen doors and gap fillers. Otherwise, the track and OH are the same, although upgraded over the years, and most of the power and signalling upgrades could be carried out while the line was still operational. So what's different?

The ATP rollout across the Sydney Trains/NSW Trains electrified network has now been completed and is operational on most sectors where rolling stock is capable and drivers are qualified. I don't recall there being any significant shutdowns for that. The next stage of ETCS Level 2 and ATO will build on the compatibility with the ATP technology and I don't expect there will be any significant disruption to existing services in the transition, unlike the introduction of a completely incompatible GoA4 system on the Bankstown Line.

The complexity on the Bankstown Line as I see it, is protecting the rail corridor from intrusion, which isn't as easy on a surface railway.

No true, your statement wasn't 100% clear it was BKS only, regardless so what? As I pointed out there are a number of other suburban lines in Australia that have been fully or partially shut down for at least 12mth as part of a major modernisation project which including, adding new lines, electricification , rebuilding grossly out dated infratructure.

Thats the point, the stations are 100 years old, so where are all the services, how well was it documented, how well as the levels, the foundations etc designed, recorded etc etc. What aspects of the orginal electrification and signalling is still used, were upgrades done to the plan at the time, when ever was it built?

Again, the BKS line had to go to enable more services on other lines and in the process connected to the only practical option, the southern end of the M1. There is no other alternative without building an entire new branch which still left BKs in the cold, its not even worth discussing anymore as its been done to death.

We are not talking just changing the signally system are we?

[Transtopic]

We are not talking just changing the signally system are we?

Certainly not. Most if not all of the station upgrades have already been completed along with other works. Track and OH have been upgraded over the decades and I doubt if much change would be needed there. I'd like to know what other "complexities" are involved when things like resignalling and power upgrades should be straight forward now that they have full possession of the rail corridor. It's strange that they are still unable to give a timeline for completion at this late stage. They should be more transparent and identify the actual issues still outstanding, rather than just a broad excuse blaming complexity and industrial action.

Just BTW, we could have had double the relief on the City Circle if they had proceeded with the original proposal for the new CBD and harbour tunnel line as an extension of the existing network from Eveleigh to Chatswood, which would have included diverting some T8 services via Sydenham to the new link. But that's ancient history now. There are still 4 of the previous 10xT3 tph services continuing to run on the City Circle, but via Town Hall instead of Museum. A net reduction of 6tph, when it could have been much more.

[tonyp]

Automated trams is interesting. I'm thinking for Melbourne type tram operations this would be a bit more challenging, however for Sydney, Brisbane, Canberra and Adelaide type operations, ie LR more than tram it would be fairly straight forward.

Do they de-man the tram completely or just move the staff member into the body of the tram checking revenue etc?

Tram automation is designed for an open street environment, pretty much like a Tesla car, so it has to anticipate any potential lateral intrusions as well as what's merely ahead of it (as with our metro). As in jurisdictions where they allow self-driving Teslas but, for now, require the car to be attended by a "driver" who can intervene if necessary, so will automated trams likely have an attendant on board. But the long-term objective is to run them unstaffed.

For the sceptics on this forum, here is a Tesla in America that is, first, summoned to drive itself from its parking spot within a supermarket car park (legal I guess because it's on private property) to pick up the driver at the supermarket entry, then to drive him home without any intervention. It gets a bit boring, so feel free to skip ahead. I'm glad I don't have his drive to the shops, it's awful but that's America for you!

https://www.youtube.com/watch?v=GXfjAAr ... L&index=20

This is pretty-much what we have to look forward to with trams.

[rtt_rules]

Automated trams is interesting. I'm thinking for Melbourne type tram operations this would be a bit more challenging, however for Sydney, Brisbane, Canberra and Adelaide type operations, ie LR more than tram it would be fairly straight forward.

Do they de-man the tram completely or just move the staff member into the body of the tram checking revenue etc?

Tram automation is designed for an open street environment, pretty much like a Tesla car, so it has to anticipate any potential lateral intrusions as well as what's merely ahead of it (as with our metro). As in jurisdictions where they allow self-driving Teslas but, for now, require the car to be attended by a "driver" who can intervene if necessary, so will automated trams likely have an attendant on board. But the long-term objective is to run them unstaffed.

For the sceptics on this forum, here is a Tesla in America that is, first, summoned to drive itself from its parking spot within a supermarket car park (legal I guess because it's on private property) to pick up the driver at the supermarket entry, then to drive him home without any intervention. It gets a bit boring, so feel free to skip ahead. I'm glad I don't have his drive to the shops, it's awful but that's America for you!

https://www.youtube.com/watch?v=GXfjAAr ... L&index=20

This is pretty-much what we have to look forward to with trams.

Thanks.
It should be fairly straight forward then as it doesn't need to steer, just stop. Plenty of video's of cars turning in front of manual driven trams ending in tears for the car driver.

Yes, I've also seen plenty of videos of Tesla's driving in shopping centre carparks to the waiting driver, especially in the rain.

Non tesla deliverying pizza driver free
https://www.youtube.com/watch?v=pX4MJLoeXH8

I think this one is a Kia
https://www.youtube.com/watch?v=5BEugKg ... p6YQ%3D%3D

[Myrtone]

Tram automation is designed for an open street environment, pretty much like a Tesla car, so it has to anticipate any potential lateral intrusions as well as what's merely ahead of it (as with our metro). As in jurisdictions where they allow self-driving Teslas but, for now, require the car to be attended by a "driver" who can intervene if necessary, so will automated trams likely have an attendant on board. But the long-term objective is to run them unstaffed.

Why not go for automated light metro? That has been in use since the 1980s, for example on the Docklands Light Railway. Off-street rail is preferable because it provides a level of mobility beyond that of road transport not at the expense of road transport, and it can also have a higher frequency than achievable with a street running service.

[rtt_rules]

Why not go for automated light metro? That has been in use since the 1980s, for example on the Docklands Light Railway. Off-street rail is preferable because it provides a level of mobility beyond that of road transport not at the expense of road transport, and it can also have a higher frequency than achievable with a street running service.

No reason not to, however is the technology there yet? Approved by regulators? But as the other poster said, it will eventually come, just like ST will one day in the future be also driverless, its just a matter of time.

[Myrtone]

Automated light metro already exists!!! I noted that and gave an example.

[rtt_rules]

Automated light metro already exists!!! I noted that and gave an example.

You havn't provided an example that is comparable with Sydney's light railway which was the general context of the topic.






If I have missed something, please feel free to correct.

[Aurora]

Operating costs for metro have been kept commercial in confidence so far, but we might get a crack at seeing them this year. Overseas experience is that unmanned automated commuter railways reduce opex by 30 to 50%.

In its 2023/24 annual report, employee costs of $1.588 billion only accounted for 34% of Sydney Trains' $4.729 billion cost base. While drivers and guards would make up a large part of this, that wouldn't be enough to generate a 30%+ cost saving.

Add another $100+ million for the managers and executives then another $350 million for critical functions now provided by TfNSW Shared Services. Gets us to $2billion and over 42% of your base.

[rtt_rules]

In its 2023/24 annual report, employee costs of $1.588 billion only accounted for 34% of Sydney Trains' $4.729 billion cost base. While drivers and guards would make up a large part of this, that wouldn't be enough to generate a 30%+ cost saving.

Add another $100+ million for the managers and executives then another $350 million for critical functions now provided by TfNSW Shared Services. Gets us to $2billion and over 42% of your base.

Reasons for automated trains lower cost
- On boards staff a minor component and many operators still have 1 person.
- Automated Metro's are often only build in high capacity routes
- Automated Metro's seem to have a greater focus on revenue protection, more secure gates, more station staff but also by design many are either UG or elevated so its harder to get out without using the gates.
- Train sets are often smaller in overall capacity so its easier to run services to match demand while still maintaining a high frequency.
- Train design is more common, less customisation costs
- Capital construction costs are lower per km as using standard design in tunneling, tunnel fit out etc and other aspects.

Which takes us to WSA Metro. The new Metro is obviously smaller train sets, but wider so both seated and standing room. How to you operate the service effectively with 8 car DD's with double manning per set? We have the same issue for T6 and T7 and formly Carilingford. So for Carlingford frequency was compromised so bad it made it unpopular. T7 is every 10min, yet when you do the passenger loadings, a L1 tram would work most of the day on 10min window. T6 will no doubt be just as bad.

So the rich man's problem for Sydney Trains is they are moving to high capacity trains, ie fixed 8 car lengths with 2 staff per train and to make this work efficently they either need volume or lower frequency, the later of which is how you loose numbers if you drop above a crital time gap which is dependent on the distance of the service.

For example Carlingford line isn't that far to T9 or T1 West, i.e. 2.5 to 5 km depending on which station to which line. So unless you are walk up distance to Carlingford station or the timetable just matches your need, why bother use a sub-standard service that involves a change to make it even longer? Quick trip to Epping for a CCN express, M1 to the city (back then via Chatswood) or T1 west at Paramatta. Its why personally I think T6 service won't last due to having only one station in the middle not part of T3 or M1 of significance.

[flitter]

What was the question again?

[Transtopic]

In its 2023/24 annual report, employee costs of $1.588 billion only accounted for 34% of Sydney Trains' $4.729 billion cost base. While drivers and guards would make up a large part of this, that wouldn't be enough to generate a 30%+ cost saving.

Add another $100+ million for the managers and executives then another $350 million for critical functions now provided by TfNSW Shared Services. Gets us to $2billion and over 42% of your base.

What are the critical functions provided by TfNSW Shared Services and are they directly related to employee costs exclusive to Sydney Trains? Would not the cost of managers and executives also be included in employee costs?

[rtt_rules]

Add another $100+ million for the managers and executives then another $350 million for critical functions now provided by TfNSW Shared Services. Gets us to $2billion and over 42% of your base.

What are the critical functions provided by TfNSW Shared Services and are they directly related to employee costs exclusive to Sydney Trains? Would not the cost of managers and executives also be included in employee costs?

It should have been labeled overhead employment costs which covers all shared services and included Senior execuative management. At Rio Tinto we were told allow roungly 75 to 100% of the salary as overheads to cover full employment costs.

[Aurora]

Add another $100+ million for the managers and executives then another $350 million for critical functions now provided by TfNSW Shared Services. Gets us to $2billion and over 42% of your base.

What are the critical functions provided by TfNSW Shared Services and are they directly related to employee costs exclusive to Sydney Trains? Would not the cost of managers and executives also be included in employee costs?

So there are a number of things to decipher - firstly, yes they are not Sydney Trains exclusive costs, but includes payroll, technology, procurement, recruitment, personnel services (corporate services) across various transport agencies. Sydney Trains would make up a significant portion of those costs with its now 15,000 plus headcount and significant physical footprint. No, managerial/executive costs are separate as that is how they have been structured.

Obviously not 100% of those costs would be saved, as a small portion of head office/managerial/frontline would be required as would much of the maintenance division - probably the one area that would be touched the least.

[Transtopic]

What are the critical functions provided by TfNSW Shared Services and are they directly related to employee costs exclusive to Sydney Trains? Would not the cost of managers and executives also be included in employee costs?

So there are a number of things to decipher - firstly, yes they are not Sydney Trains exclusive costs, but includes payroll, technology, procurement, recruitment, personnel services (corporate services) across various transport agencies. Sydney Trains would make up a significant portion of those costs with its now 15,000 plus headcount and significant physical footprint. No, managerial/executive costs are separate as that is how they have been structured.

Obviously not 100% of those costs would be saved, as a small portion of head office/managerial/frontline would be required as would much of the maintenance division - probably the one area that would be touched the least.

Thanks for that.