Tuesday, November 09, 2021

Timetable Tuesday #141: How frequently will Metro Tunnel trains run?

If I was a western suburbs MP I'd be concerned that the basic 2016 Metro Tunnel Business Case proposes a full service in the east but turns half the trains back before they get to most of the west. I'd instead want a commitment that the west will get a full Metro Tunnel service from Day One with no turnbacks outside peak times.  

What if one of those MPs wanted to check what was planned? They might first scour publicly available information to see if there was anything more recent that could allay their concerns. They might have a hard time. 

The Metro Tunnel FAQs describes some collective benefits of the Metro Tunnel. For example enabling half a million more trips per week. Victoria's Big Build discusses benefits in terms of journey times, presumably in-train. I'm not sure though if the benefits of reduced waiting are sufficiently addressed. This is despite waits being a large component of total journey time and it being perceived as longer than in-train time. 


Frequency, a key specification of any public transport system, is mentioned in the FAQ but it's vague. It's as if those charged with selling the benefits of Metro Tunnel either don't know its significance or have not been provided with information, like an operating plan, that would allow them to confidently communicate the improvements coming. 

From a user perspective an urban railway that operates every 20 minutes is very different from one running every five or ten. The former is heavily dependent on set timetables, whereas the latter is 'turn up and go'. The former may involve waiting longer than the journey length, whereas the latter rarely does. Peoples' willingness to make complex trips that involve one or more changes is highest with frequent service. 

Frequency can drive planning decisions like land use, density and parking rules. As well as household decisions like whether to buy an extra car. High frequency transforms the user experience. Whether somewhere like Sunshine or Watergardens gets a 20 or 10 minute service makes the difference between whether Metro Tunnel's impact is merely incremental or the step change that it deserves to be.  

The 2016 business case

For specific numbers one needs to go to the Metro Tunnel Project Business Case. Appendix 4 is a proposed service plan (from February 2016). The service plan is arguably the project's key document as it sets down how the tunnel will work, what capacity it will enable and its relationship with the broader transport network. Ideally it should drive what gets chosen to be built and the specifications that it is built to. 

Page 10 of this document (Table 2-2) sets down proposed 'Day 1' service levels. These include not only lines that feed the Metro Tunnel but lines that benefit by cascading capacity-enhancing upgrades. The inter and off-peak parts of these are some of the most exciting benefits of the Metro Tunnel project (even though most could have been done independently of it and sooner). 


1. Service levels are shown for the morning peak train numbers, similar numbers would apply in the PM peak.

2 Figures shown are generally representative of late evening weekday service, some lines may operate different service levels at other off-peak times (for example early morning weekday, Saturday or Sunday services). 

Peak capacity is perhaps less important now than it was two years ago. Also it is easier and cheaper to attract higher off-peak patronage than peak patronage, with usage elasticity with increased frequency higher off-peak than peak (even before COVID). Hence I will focus on Interpeak/Counter-peak and 'Typical other off-peak', ie the last two columns. 

Interpeak/counterpeak service frequency

First of all interpeak/counterpeak. I've mapped the listed lines below. The thicker the line the higher the frequency. A yellow arrow besides each section indicates a service increase. Click for a better view.

The dark blue NW-SE line shows the Metro Tunnel and its lines. The service increase here is underwhelming. At all but a handful of stations interpeak service frequencies will not be improved, according to the Business Case service plan. City to Dandenong already has 6 trains per hour. Lines from there to Pakenham and Cranbourne now have and will get 3 trains per hour. 

Four stations are the main winners. Middle Footscray and West Footscray will finally get a turn-up-and-go ten minute service, greatly improving connectivity in the city direction. Meanwhile Diggers Rest and Sunbury will get a full 20 minute service interpeak rather than the current 40 minute headway. Footscray also gets more trains, although as its service is already frequent its main gain will be much better connectivity to Parkville and the Swanston St spine and also from Newport trains (already mostly every 10 min or better). 

The biggest disappointment is the busy portion between Tottenham and Watergardens. As far as we know this will remain with their 3 trains per hour, unchanged from now. This section includes Sunshine, which has been touted as a 'super hub' and 'National Employment and Innovation Cluster'. 

The business case service plan does not support such visions for Sunshine. By turning trains back at West Footscray there is a risk that even in peak periods there could be 20 minute waits between trains for people travelling to Sunshine from the city direction. Or at least that's what it says in the 2016 business case. 

Far from uniting west and east with a simple frequent through line, the passenger experience for those coming from the east could involve a 50% chance of being booted off at West Footscray. Unless the Day 1 timetable include a 10 minute interpeak service to Watergardens, the Metro Tunnel will not live up to the hype as far as Melbourne's west is concerned. And retaining a basic off-peak train service about the same as distant Pakenham or Cranbourne get now will not make Sunshine an accessible destination attractive to set up business in. 

Some lines not on the Metro Tunnel get service gains higher than most of those that the tunnel serves. Most notable is the Craigieburn line with a proposed increase from 3 to 6 trains per hour. As a busy line interpeak this is well deserved. As is the concept that the tunnel is not just about one rail corridor, but has flow-on benefits to other parts of the network. 

Upfield has a similar gain, though there is some doubt was to whether the ten minute service will start at Gowrie or Upfield. Either way this will greatly improve connectivity in the Brunswick/Coburg area and take pressure off the 19 tram, which although slow is locally preferred over the Upfield line due to the latter's currently low frequency. Peak capacity on both it and Craigieburn is also increased as Sunbury line train paths are freed up in the City Loop.  

Running the Dandenong lines via the Metro Tunnel likewise takes them out of the City Loop. Their place in the Loop is taken by the Frankston line. That in turn creates a vacancy for lines from Newport that currently form the cross-city group to Frankston. 

With the Frankston line now otherwise committed the cross-city group will now join Sandringham and Newport. Risks for Sandringham may include less reliability given the linkage to the more complex lines in the west including the single track Altona Loop (at least on weekends). However these may be outweighed by a consistent service to Southern Cross and an increase in Sandringham train interpeak frequency from 4 to 6 trains per hour. As well as extending turn-up-and-go daytime service to dense areas like Prahran (at least on weekdays), this permits better bus connectivity for east-west routes that run between the Sandringham and Frankston lines as both will offer a basic 10 minute frequency as opposed to one being 15 and the other 10. 

What does all this mean? The number of stations with a 10 minute interpeak frequency would approximately double, with Craigieburn, Sandringham and most of Upfield joining the fray. This is a large improvement, although less than what we could have had five years ago had the Network Development Plan Metropolitan Rail been followed. Sunshine and surrounds though are entitled to feel dudded if their stations don't become part of the ten minute network on Day 1, with train turnbacks at West Footscray being seen as a snub to the west. 

Also conspicuously missing is action for the very cheap to upgrade and politically marginal Belgrave/Lilydale lines. With any luck the government will be saving this as a promise for the 2022 campaign with early delivery in 2023-24, ie before the Metro Tunnel opens. As well as being sooner this is preferable as you aren't doing too much at once and can better stagger train driver training and any recruitment. 

Night (and other off-peak) service frequency

This is described as 'typical other off-peak', with late weeknight considered representative. Weekend evening services are currently half-hourly on most lines including the busy Dandenong line. Sunday mornings often offer even less service, with 40 and even 60 minute headways often operated. The map below though sticks to weeknight service on the various lines. 

The Footscray West turnback 'Berlin wall' remains an issue at night with the double the trains running east than west. Nevertheless there's still improvement in the west, with the Sunbury line improving from 2 to 3 trains per hour. Craigieburn and Upfield would get a similar gain. 

This change will mean that instead of a 20 minute evening frequency being rare it now becomes common on just over half our lines. Indeed the only ones to miss out are those on the Clifton Hill and Burnley groups. 20 minutes is still not turn-up-and-go service but it will cut waits for connections and provide many consistently serendipitous connections with trams that also run every 20 minutes six (and occasionally 7) nights per week. As perspective, Sydney's rail network enjoys a 15 minute evening frequency on most lines, including on weekends, to suburbs much further out than Sunshine. 

What about weekend daytime services? I'm assuming these are the same as proposed for nights. If so that means little change from now in nearly all cases, including Craigieburn, Upfield and Sandringham that would get weekday interpeak-peak upgrades but not weekends. Footnote 2 below the previous table is not clear, with reference to 'different service levels'. Clarity here would be desirable, with a 7 day 10 minute service (like enjoyed by the Frankston line) far more attractive than just every 10 minutes weekday interpeak. 

Extended service program proposal

The Business Case nominated potential extra projects in an extended program. This was to future-proof Melbourne Metro and deliver more service to growth areas, chiefly Melton which would get electrified. Sunbury and Melton would both get three trains per hour off-peak with nothing terminating at West Footscray. A welcome byproduct of this would be extending the six trains per hour west to Sunshine, as shown below: 

It should be stressed that all discussion here relates to the business case from 2016. This made provision for Melton but not airport rail. And even under the Extended Program it leaves the key Albion - Watergardens section with just a 20 minute service despite strong existing interpeak patronage, favourable demographics and development around Albion, Ginifer and St Albans as part of the NEIC.

Since then priorities have switched. Last year the premier announced airport rail via Sunshine. This would run every 10 minutes and be routed via the Metro Tunnel to Dandenong line suburbs. Any update to the service plan would have to take this announcement into account, along with other proposals, eg for faster trains to Geelong, what's happening at Wyndham Vale and the Melton corridor. 

Historical background

The abovementioned 2016 business case is not the first plan to overlook frequent service on most of the Sunbury line. 2013's Network Development Plan - Metropolitan Rail proposed 10 minute off-peak service on lines such as Craigieburn, South Morang and the Sunbury line as far as Sunshine. Craigieburn and South Morang are both about 30 km from the CBD, similar to Watergardens that, in contrast, would have missed out. In support of that plan, at least Sunshine would have got an off-peak boost, with this (and many other service upgrades) by 2016. Unfortunately not even this happened, with minister Allan flicking the switch from service to infrastructure, badly neglecting the former. 

Consequently documents released from both sides politics give the Albion - Watergardens portion short shrift, proposing only slighly better service than runs now. Part may be due to representational weakness on both sides of politics with the area being considered safe for Labor and unwinnable for the Liberals (although the otherwise strong 2018 election result did have some warning signs for the ALP in some western areas).

At least one local Labor MP (Marlene Kairouz) has been named in the IBAC Operation Watts hearings with regard to party branch stacking and misuse of taxpayers funds with regards staffing, while the Liberals' former transport spokesman David Davis has not vigorously pressed service issues, despite these being a weakness for Labor that could have been capitalised on. 

Hence both major sides might have been either distracted or uninterested to pursue these matters. The post 2015 transport hierarchy could have been weaker on service matters than that a few years prior. This might have led to insufficient oversight when preparing both the 2016 SRL Business Case and even the 'Extended Service Program' to ensure that the Sunbury line (at least to Watergardens) had a turn-up-and-go service specified from Day One of the Metro Tunnel.


The Metro Tunnel proposes large and welcome weekday peak and interpeak train service improvements for Craigieburn, Sandringham and part of the Upfield line assuming that the 2016 Business Case plan is followed. Some evening upgrades are also proposed. 

However only running half a service beyond West Footscray risks making the Metro Tunnel a bittersweet project for Melbourne's west between Tottenham and Watergardens. The service plan in the 2016 business case, if followed, could cause Melbourne's biggest public transport project in a generation to perpetuate, not fix, inequalities between the west and the even more accessible south-east.  

Service frequency is every bit as important as infrastructure in shaping how many people benefit from the transport network. The 'infrastructure first' emphasis of the current government and the Department of Transport's inability to produce an integrated transport plan (that would include a multimodal coordination framework) sidelined service issues for several years. Hopefully we are seeing a revival of interest in service after some lean years, with encouraging signs including train and bus timetable overhauls in 2021 along with Victoria's Bus Plan.   

The Brimbank community needs to know what service they're getting when the Metro Tunnel opens. This could address concerns sparked by the meagre offerings proposed in the 2016 Business Case. 

It could be helpful for the government to release a revised service plan that features all-day/7 day frequent and seamless east - west connectivity between at least Watergardens and Dandenong. That way the west gets the frequent service that its population, activity and demographics deserve and Metro Tunnel benefits are maximised.  

Index to Timetable Tuesday items here


Anonymous said...

Wasn't there a plan to stop the weekend shuttles on the Williamstown line?

Peter Parker said...

Yes. The 2021 timetable made inroads towards this during the day. But still shuttles early am and at night. https://melbourneontransit.blogspot.com/2021/01/building-melbournes-useful-network-part_8.html

Steve Gelsi said...

Hopefully the West Footscray turnback quickly becomes a white elephant with Airport rail and electrification to Melton and Wyndham Vale. With four destinations beyond Sunshine, you'd eventually get a minimum 15 trains per hour stopping at Sunshine (assuming 10 minutes on the Airport line and 20 minutes to Sunbury, Melton and Wyndham Vale). Of course, that's still terrible frequency beyond Sunshine. As an interim solution for Sunshine (unfortunately not for Tottenham, as there are no platforms) could V/Line trains outside the peak stop at Sunshine without restrictions?