Melbourne orbital rail proposal

Okay, I didn’t see this coming.

Orbital rail for the middle ring of Melbourne, announced on Facebook this morning.

Labor say they want to build a 90km suburban rail loop from Cheltenham (by which they appear to mean Southland), Clayton, Monash, Glen Waverley, Deakin Burwood, Box Hill, Doncaster, Heidelberg, Latrobe Uni, Reservoir, Fawkner, Broadmeadows, Airport, Sunshine, to Werribee.

Some of the detail is a little hazy, and would be subject to further study, but it seems to be aimed at hitting the biggest suburban traffic generators that don’t currently have rail.

It’s unclear where (or if) intermediate stops might be located, particularly between the Airport and Werribee.

At least from Cheltenham to the Airport would be underground tunnels. The whole project would support a more polycentric city, which makes a lot of sense, enabling cross-suburban trips in a way that Smartbuses don’t quite manage, and help Melbourne develop more jobs not just in its CBD and inner suburbs, but across the greater city.

Labor orbital rail plan

Train services on both the new and existing lines would need to be frequent of course, with easy connections to and from the existing network. It would also provide regional train connections at Clayton, Broadmeadows and Sunshine.

Labor say the line would not only be Melbourne’s longest, but also Melbourne’s busiest, with around 400,000 passenger trips per day.

The project might divert some people away from the busiest inner-city part of the rail network, but would inevitably generate more public transport trips as well.

Doncaster would get rail – just not in the way most people expected… Doncaster to the City via Box Hill would become a more time-competitive option.

I wonder if they assume the orbital trains would be similar technology to the existing fleet, or something different? To support frequent services, perhaps they would build the stations for full-length trains, but initially run shorter ones.

Business case funding of $300 million will be provided if Labor is re-elected, then construction starting 2022, in a phased approach but most likely concentrating initially on the southeastern section first, with the Airport to Sunshine portion falling under the existing Airport rail project, which would share the same alignment.

And yes, suddenly some other Labor ideas make sense in context, such as Monash Uni/Rowville light rail instead of heavy — the heavy rail option would be Clayton to Monash instead.

This is Big City thinking, and should be applauded, though who knows how much it will cost and how long it’ll take to build. Billions and decades, I would expect.

It’s incredibly ambitious, which is great to see. Not that it lets them off the hook on other, smaller things like fixing Melbourne’s buses (particularly to help people access railway stations) and all-day frequent trains on the existing lines.

But if they can pull it off, it’s huge, a game changer for the city.

By Daniel Bowen

Transport blogger / campaigner and spokesperson for the Public Transport Users Association / professional geek.
Bunurong land, Melbourne, Australia.
Opinions on this blog are all mine.

47 replies on “Melbourne orbital rail proposal”

Great idea. It’s exactly what many people have been wanting for many years and it would revolutionise Melbourne. It will be interesting to see how opponents frame their argument.

My hope is that the Sunshine to Werribee Section is above ground, and goes via Wyndham Vale (which would make it a loop shape), but we will see I guess.

It will be interesting to see what this will mean for other proposed rail projects.

I would assume a Doncaster Line is off the table. This would then make the business case for Metro 2 quite difficult. There isn’t sufficient demand at the Western end, and at the Northern end the demand could be more cost effectively managed with high capacity signalling between Clifton Hill and Jolimont, grade separation of the Clifton Hill Junction and running the new HCMT on the Mernda line

Managing capacity on the Mernda/Hurstbridge lines will be a tough proposition without Metro 2 – in the long term. It is true that high capacity signalling and grade separation will help, but the corridor is already quite constrained, there is also the problem of platform congestion and delay mitigation – which means headways are limited. There is only one platform at extremely busy stops like Flinders St and Melbourne Central (which will have increasing numbers interchanging) and there is no buffer time to reduce the effect of delays from when trains enter Clifton Hill, to when they leave past Clifton Hill again. The corridor is pretty awful in general anyway – with a 55km/h maximum speed, and many stops less than 500m apart, which would get even more ridiculous if platforms are lengthened. It is very space constrained, so track amplification would be exceptionally costly.

There is also huge expected patronage, wih growth of Wollert and other areas like that, and if the lines are connected through LaTrobe and the major hub that is supposedly being planned for the area.

Metro 2 is a hugely expensive project, but it does not seem like there any other particularly viable alternative. It would be a lot easier if there was – but it is a very constrained area.

Great idea, but it needs to start at Sandringham (only an extra 4.8 km), and then it would link most of Melbourne’s above ground rail lines.

It would be best if, they extended the Alamein line to Oakliegh in the short term. They could have done that as a skyrail, with the equipment being in place at Murrumbeena, and done at a reasonable cost. Direct links between Oakliegh, East Malvern and Camberwell.

I am supportive of this new idea. Although must it be wholly underground?

Would it be cheaper and faster if they built it above ground alongside EastLink and then joined it in with the new North East link?
I mean I am all for a cross city transport link, but come on, its an election year, and the cost of this is so extreme we can’t afford 50 km of tunnel never mind 90km.
Be cheaper and faster to complete if they simply bought houses and built an auto-bahn bus track like Adelaide has. And a lot faster for passengers too.

Great project, but it needs to start at Sandringham, with a potential additional stop near the industrial area/new large apartments on Bay Rd

The most appealing aspect of this is getting rail to the largest employment clusters that currently don’t – particularly the airport (already on the table), universities, shopping centres and industrial hubs. None of these areas are going to get direct radial access to the city. An orbital line will need to complement the radial network, meaning minimal interchange times, but would be more competitive to private transport than the current orbital bus options that run via these employment clusters. I would think that it might be competitive with some existing cross-city rail trips, depending on interchange in the city. I find it amazing that I can get from home in Footscray to Deakin Waterfront in Geelong (and possibly to Waurn Ponds as well) quicker than to Deakin Burwood.

I wonder what mode they will choose for this line. I’m personally hoping it will be a real metro/rapid transit with a very high average speed so that it can effectively serve metropolitan area and not take forever. I reckon it should be about 1h15m end to end or an average speed of 72km/h, similar in speed to Line 18 of Paris Metro (35km in 30 minutes, 70 km/h). If it’s a real metro then I could imagine it being independent and also fully automated.

Yes Ricky,
Everyone wants it to start where they live and go to where they work or study. So clearly it should not start at Sandringham. It should start at Monash Uni, and go to Rowville first. :)

Perth has had orbital rail on the books for a while, with segments of it presently under construction or in planning. Sydney has the Cumberland line and the general interconnectivity of the lines has enabled some cross-regional services that have come and gone over the years.

Great initiative and long term vision
It also needs to
a) be opened in 10 years
b) add a line from Frankston to Cranbourne, Clyde, Narre Warren, Springvale, Glen Waverley. The City of Casey needs infrastructure to support the ongoing growth, no transport in the Berwick Waters Estate which has been opened for 4 years, I need to walk 32 minutes to a bus or 1 hr 7 mins to a train.
The City of Casey is 40km south east of Melbourne. It is Victoria’s largest and fastest-growing municipality and the dominant growth area for south-eastern Melbourne.
Casey is also the third-fastest growing council in Australia behind the Gold Coast and Brisbane City Councils.

If this really goes through, it’ll be pretty much needed.

I wonder if they’ll run smaller trainsets. In Singapore, the suburban lines that don’t or barely touch the CBD only run 3 or 4-car trainsets. I would favour frequency over capacity.

This is an extremely good idea, and very exciting.

Prediction: Underground orbital rail link with construction starting in 2022 becomes a link from Cheltenham to Monash, largely as a skyrail, construction starting in 2030. Orbit finally complete by 2080.

This feels too far out; by connecting ends of lines that get her it doesn’t do very much for intra-suburb transport closer in. Preston to Coburg or Northcote to Moonee Ponds, two examples among many, would not be helped by this. I’m all for grand visions like this but it’s seems like this should form part of an integrated strategy that answer questions like this, along with metro rail 2, airport rail, buses, trams etc.

It’s been a long time since he last one but it would help evaluate this idea in context.

An visionary and ambitious project to link disconnected major parts of the city by rail. This plan seems to be in the making for a while (as you point out) and the preferred rail alignment between Sunshine and Airport, and the proposed Caulfield-Monash-Rowville tram line makes much more sense.

Still neglects many suburbs already wanting. The big black hole between Glen Waverley and Ferntree Gully still misses out altogether. Light rail from Monash to Rowville won’t cut it – Rowville is still quite a distance from Wantirna. What about the Knox City and Caribbean commercial hubs already there and growing? The whole area is already full, unlike Mernda which appears to still be green paddocks. Location of the line needs some discussion and tweaking.

It great to see that the outer circle is about to get reincarnation in underground form (yes, I realise the corridor of this proposal is going to suburbs that have never had rail). I wonder if the tracks of the new line will be physically fed into existing rail tracks?

I can think of at least one example of where doing so would make sense. If the Geelong line is electrified in its entirety, it should at least be electrified as far out as Lara and while they’re at it, build a rail branch at Lara to Avalon Airport and feed all trains from Avalon through the proposed underground loop (via Lara, Little River and Werribee) and all other trains originating at Geelong can still go via RRL as they do now.

While thinking of “ring lines” it would be easy and relatively cheap to extend the Alamein line on the surface to the Pakenham/Cranbourne line at the south end. and extend it through to the Hurstbridge line on the north end. And it would cross the Belgrave/Lilydale lines at East Camberwell. The land is already there and waiting. So easy that they could have thought of this 120 years ago. Oh wait, they did.

Bringing down the cost of tunneling will need to be one of the initial steps in this project. Our present Metro 1 is among the most expensive in the world at $1b per kilometer. Perth is constructing its airport line underground for $200m per kilometer. Differences are firstly that Metro 1 is built through legacy areas with a lot of existing infrastructure, such as Swanston St, whereas Perth’s is mainly brownfield areas. Secondly, Perth already had some recent tunneling experience with the CBD section of its Mandurah line.

The public sector needs to lead on this, and not expect private construction contractors to take large financial risks or develop innovative methods. Public funding for pilot projects, overseas learning visits, and a good long pipeline of projects, should all be used to develop a project that delivers value to the Victorian public.

It would be a pity if this project delayed Metro 2. One of the strengths of Melbourne over the last 20 years has been the development of Docklands as an extension of the CBD. Employees are more productive in dense employment areas, and the CBD is accessible from all suburbs leading to a better field of applicants for positions. Yet how many more years of development are there before Docklands fills up? We need Fisherman’s Bend as the next central high-density employment area. But it won’t work with a slow tram. It needs Metro 2.

Much of this can be overground:

The railway from Sunshine to Tullamarine can be elevated – either above the existing goods line or the existing M80. The M80 also goes right past Broadmeadows Station – so the railway from Tullamarine to Broadmeadows can be entirely overground.

There is even a massive and empty corridor from runway 27 (airport) to Pascoe Vale Rd (Broadmeadows) – just look up “Chandos Place, Attwood” on Google Maps Satellite View and the corridor is parallel to it.

The M80 actually continues to Bundoora. So the railway can be overground till Bundoora.

It was madness to build an apartment block at the end of the Glen Waverley line – they should have extended the Glen Waverley line by 200 metres so that it terminates right on Springvale Rd – just like Nunawading Station. Then they could have converted 2 lanes of the 6-lane Springvale Rd into a busway. Now, the 902/901 bus has to take a detour to get to Glen Waverley Station!

They should convert 2 lanes of EastLink into a railway to allow people to get from Ringwood to Dandenong.

I noticed a missing connection, that being between Sandringham and Chelthenham. How will the government fill this missing link to cater for the residents along the Sandringham line, Black Rock and Beaumaris? I’m guessing it’s not heavy rail because it’s risky to build underground rail tunnels close to the coast.

Pretty interesting proposal. Had a look and other (big) cities have them, but they take a long time to develop. Interesting there seems to be no more than two loops. Tokyo had a proposal for one between the inner loop (Yamanote) and the mega (Musashino) loop but its on the 25 year plan :). Probably want to think about freight along the line too.

I watched that stupid Glen Waverley apartment block being built and my thoughts were exactly the same as yours, still are. The Glen Waverley line originally terminated right at Springvale Rd but they moved it back 200 metres to provide a car park for the shoppers. In my opinion the Glen Waverley line should have been extended much more than 200 metres, like about 13km to link up with the Belgrave line. That would have gone some way to filling in the transport black hole in Knox, which still won’t be filled under this new plan.

On paper sounds like a good idea. BUT I am suspicious, where did it come from? Who is behind it? Any academics/experts been involved or is it a case of ‘Transurban Rail’ /private sector putting it up?? A la the Westgate Tunnel Project??
AND does it mean that MM2 is off the table? because maybe Transurban doesn’t want anything that would reduce the attraction of the Tunnel???? I

1. Use Sunshine – Newport – Williamston Pier, instead of Sunshine – Werribee. This uses uses an existing corridor.

2. Monash – Oakleigh – Moorabbin – Brighton Beach instead of Monash – Clayton – Cheltenham . Cheltenham is already well serviced by Cheltenam AND Southland Stations.

3. Extend Alamain to Oakliegh

4. Use the spoil from the tunnel to build a Ferry Terminal Interchange Station offshore at Williamstown Pier and a Ferry Terminal Interchange Station offshore at Brighton Beach to complete the circle.

EVERY radial rail line now links to Melbourne Airport and other major trip generators with one interchange outside of the CBD.

I don’t see the point of building a new direct rail line from Sunshine to Werribee as we already have the Regional Rail Link that provides this connection via Tarneit, once the missing section between Wyndham Vale and Werribee is constructed.

Pleased to see Broadmeadows (and thus northern Victoria, via a transfer from V/Line to Metro at Broadmeadows) get a rail link to the airport as part of this. Of all the discussions held on airport rail, it always seemed short-sighted to look at one city-bound line only.

A shame that so much might need to be tunnelled – increasing cost, reducing ability to re-route trains on or off this line during maintenance, and just having less amenity for passengers that a view – it doesn’t seem like there would be many alternatives for much of the route, given lack of land put aside for such a project.

A great idea, good to see a similar level of ambition for public transport we’ve seen for decades with roads. But it won’t work without frequency improvements across the network. It’d be a massive waste of $50B+ to build it only to operate at 20 minute frequencies.

One issue with the proposal (possibly too early to know): if the Sunshine option for airport rail gets up, will the airport rail and the new circle line be separate tracks/services or will passengers have to transfer at Sunshine to get to/from the city?


That line between Wyndham and Sunshine for the RRL is meant for rail to and from Geelong (and partially Ballarat). Reintroducing suburban traffic into it will bring back the congestion that was seen prior to the RRL.

Hi Peter
It is not the apartment block that is the problem They would have needed grade separation even if they extended under through the car park. And on the other side of the road, there are already houses in what was the rail reservation. And there is a confluence of services and flood mitigation underneath that car park.
If they were to go underground, and the say through to Rowville ( as was one alternate Rowville line proposal, they would an alignment under what is now the carpark/busstation/ north railway road takes into account the natural fall for construction on the other side.proposal). But that ship has sailed, and it is not part of the current proposal

There are times when public-private partnerships do not make sense- because governments can always borrow more cheaply than the private sector.
There are times that they do because of construction risk. And there are times when they do because of amenity and budget crowd out.
What is being proposed as a 30+ year build. We could assume that it will have a 100 year life expectancy. Similar projects in PPP are “routinely” build in less than 12 years, with a 30 year Build Own Operate Contract , Return to Government ownership after that- with 70 years of usable life. It is done this way because of government budget impact. But also because spending at that intensity on one project crowds out public sector spending. AND it bring forward amenity by 20+ years. So yes let’s do this.with a PPP now.

I’ve read many of the comments on here and also on newspapers sites and thought I would just help to clarify some of (what I perceive) the misconceptions of this project.

Apologies for the length!

Many people are getting hung up on this being a rail project to connect rail lines. While that is true, it isn’t what the project is all about. The connection with rail lines helps to change our current radial rail network (a single hub with spokes) to a polycentric network (multiple hubs with spokes).

But the true logic lies with connecting employment clusters with the wider transport network and with each other. This is why the idea of connecting the line with the Sandringham Line or Williamstown Line, moving it further out to the ends of the lines, inward towards the city or following easier alignments like freeways are illogical because they are missing the point.

This is what it is connecting:

Education clusters (figures are estimates as campus breakdowns aren’t always available):
Monash Clayton – over 30,000 students and over 10,000 staff
Deakin Burwood – over 28,000 students and over 3,000 staff
La Trobe – over 36,000 students and over 4,000 staff

Major Retail Clusters:
Chadstone (potential 10 min future tram connection?)
The Glen
Box Hill
Pacific Werribee (potentially)

Minor Retail Clusters:

Major Medical Clusters:
Monash Medical Centre
Box Hill Hospital
Werribee Mercy (potentially)

Major Employment Clusters:
All of the above plus:
Melbourne Airport – Over 16,000 employees
East Werribee Employment Precinct (potentially)

By opening up a new larger catchment of employees or customers that don’t need to drive, the intensification of these areas will increase dramatically.

Also, by creating regional rail hubs further away from the CBD, there is the potential to reduce regional trains going all the way into the city – this could be especially important in freeing up space on the Pakenham/Cranbourne line.

While the east section of this line follows a certain logic, the western section is hazier around Werribee. With the video stating it terminates “on the Werribee line” even though the map released makes this appear to be Werribee station (same thing with Cheltenham/Southland)

The logical alignment would include the East Werribee Employment Precinct – a 775-hectare development that will house tens of thousands of new residents, students and jobs (depending on which prediction you read).

This could bring the terminus of this new line down through to Point Cook.
-Connect Sunshine station to Tarneit station (duplicate RRL)
-Underground to Pacific Werribee
-Underground to the planned East Werribee station on the Werribee Line at Derrimut Rd
-Potentially a new station within the 775-hectare site
-Underground along Sneydes or Dunnings Road to terminate in Point Cook.

While this will be large increase in stop densities compared to other sections, it would: (1) connect a suburb without rail, (2) a new high-density development (that includes a hospital and 2 university branch campuses) and (3) a major shopping centre. As this occurs at the end of the proposed line, the density of stops would not affect the overall performance of the line by increasing travel times except for start of journey users.

@Jacob (Tue 28 August 2018 at 8:06 pm), the proposal for a station at Bundoora is at La Trobe Uni. This is about 4kms south of the M80, along Plenty Rd.
Also, the M80 hardly “goes right past” Broadmeadows Station. Jacana is the Craigieburn Line station right on the Ring Road. Any future Campbellfield Station on the Upfield Line might be located very close to the Ring Road. But you have a good point about the potential for the Attwood Corridor.

@ Urban-Planning-Student says:
Fri 31 August 2018 at 11:19 am

Wow, now that you’ve mentioned it, this orbital could probably be extended to Werribee South and/or Point Cook.

Also Werribee Mercy is probably closer to Hoppers Crossing station (on a separate topic, that crossing close to it could really use with grade separation).

Top-notch analysis Urban Planning Student. Although I like the idea of a transport hub at Sunshine the line gets tortured at that point. Why not let the IFM consortium build Southern Cross- Sunshine-Airport for the air-commuters, and have the SRL instead connect from Airport to Watergardens, Caroline Springs West, Tarneit (going to be huge by the time they get to it), Hoppers Crossing/East Werribee, Point Cook. A key benefit would be access to the infrastructure precinct near to Tarneit, as well as allowing airport workers to access the airport.
P.S. Does anyone know how Cheltenham to Airport (~56km) can be done in 45 minutes with 12 stops as “modelling shows” per Daniel Andrews? Average speed in transit would be 100kph!

On your ps. In Melbourne everyone thinks in terms of the current speeds on poor quality track and poor signalling. This will be a dedicated purpose built line with stations 4 to 5 km apart. For example the Hitachi A series variants have maximum speed of 125km to 240km depending on track and purpose. The AT100 high density urban and the AT200 suburban regional intercity both have max of 160km. I think maximum speed is limited by the network and track. ( There are other competitors). So 100km average should be possible.

The ideas great but lets do it ‘piece meal’ ie build a section between 2 main towns, then run the light rail & gradually join the mating section so parts of the line can run before the 10 year period. This will aid the ever growing population and create industry because of the new transport growth as the line is being built.

7 run the rail line

Comments are closed.