UN 205: Will GPT replace DTP? An AI derived bus network for Box Hill

An ABC item earlier this month says that AI-based large language models like Chat GPT could help plan cities. An Age article says that AI bots could help slim down the Victorian public service. That's topical given the state government is looking for savings with its Silver review due to report this month. 

So for fun I gave ChatGPT a task:  

Describe a reformed bus network for the Box Hill area with a list of main routes and the roads they travel on.

This is the response, pasted verbatim, including formatting. 

🚌 Proposed Key Bus Routes for Box Hill Area

---

Route 1: Box Hill – Deakin University – Burwood Station (University Link)

• Primary Purpose: High-frequency link to Deakin University

• Roads: Whitehorse Rd → Elgar Rd → Riversdale Rd → Station St (Burwood)

• Frequency: Every 10 min peak / 15 min off-peak

---

Route 2: Box Hill – Doncaster Shoppingtown via Station St

• Primary Purpose: Cross-suburban shopping and employment access

• Roads: Station St → Doncaster Rd

• Frequency: Every 10–15 min

---

Route 3: Box Hill – Blackburn – Forest Hill – Vermont South

• Primary Purpose: East–west connector through middle suburbs

• Roads: Whitehorse Rd → Blackburn Rd → Canterbury Rd → Burwood Hwy

• Frequency: Every 15 min

---

Route 4: Box Hill – Surrey Hills – Camberwell

• Primary Purpose: Local access and feeder to Camberwell Station

• Roads: Whitehorse Rd → Union Rd → Riversdale Rd

• Frequency: Every 20 min

---

Route 5: Box Hill – Mitcham – Ringwood

• Primary Purpose: East–west orbital route

• Roads: Whitehorse Rd → Springvale Rd → Maroondah Hwy

• Frequency: Every 15 min

---

Route 6: Box Hill – Donvale – Ringwood North

• Primary Purpose: North-east local connector

• Roads: Elgar Rd → Wetherby Rd → Mitcham Rd → Warrandyte Rd

• Frequency: Every 30 min (coverage route)

---

Route 7: Box Hill – Glen Waverley

• Primary Purpose: South-east orbital route, university + train link

• Roads: Elgar Rd → Highbury Rd → Springvale Rd → Kingsway (Glen Waverley)

• Frequency: Every 15 min

---

Route 8: Box Hill – Balwyn – Kew – Clifton Hill

• Primary Purpose: North–south route into inner suburbs

• Roads: Whitehorse Rd → Balwyn Rd → Harp Rd → High St

• Frequency: Every 20 min

---

🚏 Supporting Feeder & Local Routes

• Shorter, less frequent routes feeding into Box Hill and key train stations.

• Potential local shuttles to areas like Greensborough, Mont Albert North, Forest Hill Chase, etc.

---

🎯 Network Features

• Simplified, straight routes with fewer deviations.

• Clockface frequencies for ease of use.

• Improved weekend & evening services.

• Integration with trains and trams, especially at Box Hill, Glen Waverley, and Camberwell.

• Direct access to major trip generators like Deakin University, Box Hill Hospital, and shopping centres.

Chat GPT's map

You could request a map. The output gave me a Python script. I wanted to bypass this for an image I could present directly. That required me to log in to ChatGPT. 

The result was what you see below (click for a better view). There is no geographical accuracy. Eg Ringwood North is shown south of Camberwell and a long way from Ringwood while Glen Waverley is north-east of Doncaster. 

My map of 8 main routes

Using the written route descriptions I tried to make a map that was geographically accurate with regards to Box Hill, route termini and streets the routes would run along. That proved a challenge as some route descriptions lacked sufficient detail to make their paths unambiguous. My best attempt is below:  

As instructed the network is centred on Box Hill. Some routes are sensible. Others are sensible but could be longer to connect with other destinations (like existing routes currently do). Others you might not do.

Quick comments on each are as follows: 

1. Box Hill - Burwood station. Alignment is good between Box Hill and Deakin Uni, being similar to existing popular routes. However Burwood station on the quiet Alamein line is a weak terminus. Also this section of the route would overlap the longer and more frequent 75 tram so you wouldn't' do it. .  

2. Box Hill - Doncaster. An useful and direct connection this segment is currently served by the much longer 903 orbital. If it was linked to the Route 1 proposed above it would match a longer section of the 903. 

3. Box Hill - Vermont South. Has similarities with the existing 732 but is shorter and goes along more of Whitehorse Rd. 

4. Box Hill - Camberwell. A route with benefits for Surrey Hills but it inefficiently overlaps too much of the 70 tram on Riversdale Rd. 

5. Box Hill - Ringwood. Description here is ambiguous (with mention of Springvale Rd) but I've understood it to mean a direct Whitehorse Rd/Maroondah Hwy route. Not without merit despite it paralleling a rail line as there are a large number of destinations along the highway.  Existing routes like 271, 279 and 901 serve this section but not continuously. There may be a case to be splitting Route 901 at Ringwood and South Morang and rethinking what happens in between. That might include a Box Hill - Ringwood route on the highway. 

6. Box Hill - Ringwood North. The route description was not clear so what you see on the map is possibly not what was intended. This was described as a coverage route with a 30 min frequency (most of the others were every 15 min).

7. Box Hill - Glen Waverley. The alignment and the frequency is good in that, combined with Route 1, this would provide a 7.5 minute frequency between Box Hill station and Deakin University. The current network does not provide a direct Box Hill to Glen Waverley bus as one would not add a lot of new coverage. However it does link two major activity centres. This concept has similarities with a Liberal election promise in 2022 which may be where ChatGPT got the idea. 

8. Box Hill - Clifton Hill. Alignment again not clear, especially at Clifton Hill end. Parts have similarities with existing Route 302 (though this goes into the city). 

The above was the first network option that came up. Had I asked ChatGPT a different question or given it guidance the result will have been different. 

Conclusion

This (free) version of ChatGPT AI has a way to go when it comes to designing bus networks.

But more advanced paid types with more data input and other guidance may produce more sensible networks. 

Or, given the planning and scheduling software already used in the industry like HASTUS and Remix maybe an AI plug-in to enhance its planning function is how this will evolve in the future? 

What about other applications for AI in transport? AI can give reasonable answers for things like customer feedback. Maybe even journey planning. It might help with advocacy or submission writing, like I tried here.

However the AI you get when you do a Google search can mix things up. For instance when I did various searches on the Bus Plan the result conflated official and unofficial sources, such as from DTP and my blog. Personally I would love the day that a lazy official, unaware of this, produces a report that includes slabs of my blog as official policy. 

The old saying is 'Garbage In, Garbage Out'. Proper use of AI possibly means that its search be limited to official and verified sources rather than what just anyone says. Though DTP/PTV data can be wrong so an official source is not by itself a guarantee of correctness. AI risks just exacerbating existing errors. AI also needs a knowledgeable driver - that is people using it need to be learned enough that they can spot an error and correct.  

Comments on AI's potential role in transport are appreciated and can be left below. 

See other Useful Network items here

VAGO to audit Victoria's Bus Plan

The Victorian Auditor-General has included a performance audit of Victoria's bus planning in his 2025-26 annual plan tabled in state parliament last week. VAGO's review will examine the performance of Victoria's bus services, including if the Bus Plan is on track to achieve its intended objectives and targets.

Other transport topics to be examined include myki ticket modernisation, the progress of major projects, road maintenance and customer service of outsourced Vicroads functions like driver licensing. A follow-up of the audit on integrated transport planning is also pencilled in for 2027-28. 

An audit is timely

Back to buses. An auditor-general review of bus services is both desirable and due. One was proposed six years ago but got put off. However, with it being nearly 1500 days since Victoria's Bus Plan was announced sufficient time has elapsed for an auditor to form an opinion on its effectiveness or otherwise.

Not only that but the dollar amounts involved are, in auditor-speak, material. Victoria spends the better part of $1 billion each year to run bus services, mostly through payments to private bus operators. Unlike train and tram contracts (which appear to have got more generous over the last 20 years) payments to bus operators per service kilometre delivered appear to have been fairly constant relative to CPI.

Just because we're getting a reasonable amount of service kilometres per dollar for bus does not mean that these bus and driver resources are optimally deployed to be useful to the most number of people for the most number of trips. The key determinant of this is how well routes and timetables are planned and meet the public's travel needs. 

With routes and timetables for maybe two-thirds of Melbourne's bus network substantially unchanged for 15 - 40 years, network reform (the first priority of the Bus Plan) has a far greater bearing on the value we get from buses than certain other initiatives (eg fleet electrification that may have merit for other reasons). The 'health checks' I've done on the bus network are here and here, with only a slow rate of improvement between them. 

The extent and quality of bus network reform can be difficult to measure but is easy to obfuscate. This makes it vital for an independent party like the Auditor-General to properly examine the performance record here.  

link to count-up Bus Plan clock

Audit role

What might the audit cover? VAGO audits don't always cover everything some would like. It's worth taking a step back to understand why this is.   

Different independent institutions established by parliament to monitor the executive have different  roles and emphases as set out briefly below.

The deliberations of parliamentary committees like Parliament's Accounts and Estimates Committee are shaped by the politicians on them. The Independent Broad-based Anti-Corruption Commission (IBAC) is a specialist corruption investigator with quasi-judicial powers. The State Ombudsman is a 'last resort' complaints investigator, although they may also comment on policy, especially where it may be improved. Infrastructure Victoria is a major giver of advice on transport policy (including through the media where it is more open than DTP) but this does not extend to cases where the policy has been adopted by government. Thus you won't find IV commenting about the merits or otherwise of the Suburban Rail Loop (a project of this government). Indeed their reports tend to avoid even mentioning the SRL unless necessary. 

Then there's the Auditor-General, the oldest government accountability office dating way back to Victorian self-government in 1851. Being independent and being seen to be independent is central to how auditors general do their work. They are in the camp of institutions who do not analyse or comment on government policy. But they are very interested in it being well carried out.  

The pitfalls of those who are not ministers commenting on policy are explained here

 

The difference between the policy of administration (a political responsibility) and the administration of policy (a job for bureaucrats to do and auditors to review) is explained here

Potential audit topics and departmental responses

When asked about their record, the government and DTP can (and likely will) rattle off a long list of bus service upgrades that it attributes to Victoria's Bus Plan. They represent, with few if any exceptions, progress.

But something being good might not entirely satisfy a probing auditor if the methods used to determine priorities have not been robust. 

I have suggested that past departmental secretaries might have used counting methods of doubtful rigour to exaggerate their achievements in bus network reform when under scrutiny. Including from PAEC, which is the same committee the Auditor-General reports to. I would expect that the appropriateness of performance metrics would feature in a performance auditor's test program. 

An audit might also be interested in monitoring operational characteristics such as bus route patronage and acting on the data received to optimise resourcing. An examination will likely find that such data is collected (though one might query its accuracy). The department could give legitimate examples where resources have been transferred between routes (mostly involving Transdev/Kinetic routes) for an overall greater good.

Although they will probably cite new bus contracts as making this easier, DTP might find it harder to argue that this is being done in a particularly systematic way (eg tackling low productivity and/or duplicative routes first) even though the Bus Plan flagged resourcing/patronage productivity mismatches as an issue. 

Auditors (especially) hate things being funded without an implementation plan. Even if what gets funded is sensible, likely consistent with what a plan would recommend and proves successful in practice.

They're not fans of 'bait and switch' either. That is if a department would promise something (as being consistent with the plan), not continue with it but switch to doing something else. The party being audited may claim that that something else was consistent with the plan. Or they could just write a plan that is so vague that almost anything could count as being consistent with it, with the real detail coming later.  

DTP may have left itself exposed here as it:
(a) failed to produce the Bus Plan's promised Bus Reform Implementation Plan by its 2023 deadline (with a weak answer from then Secretary Younis when quizzed here),
(b) apparently stalled on the North, North-East and Mildura bus reviews promised before the 2022 state election, and
(c) ran trials that either had no apparent progress (eg Rapid Running being extended to more routes) or were reasonably foreseeable duds (eg FlexiRide) that distracted attention from beneficial bus network reform or service upgrades (such as Greensborough will finally get after flip-flopping on FlexiRide).   

Scale is important when planning transport in a big city. One or two little route upgrades might be successful but you need to replicate it across hundreds to have a metropolitan-wide effect. Has bus plan facilitated reform to the scale needed and that it aspired to? The record so far is that it has not, with the pace of reform both slower than other cities (eg Perth) and our own record in the 2006-2010 period. Achieving the aspired pace of reform requires changes within DTP as current processes have more in common with cottage industries than the mass production needed for large-scale delivery.  

The government releasing the Bus Plan without substantial funding in 2021 and the department's inability to argue its case for this in the 2023 and 2024 state budgets wouldn't have helped its standing either. Rather than being a substantive program or project (like the WGT or SRL), the Bus Plan has been demoted to be more a thinking approach to be applied if or when the government wants to do things with buses.

Essentially the Bus Plan has been an unloved orphan for most of its first four years. There are welcome signs of revived government interest in buses in the 2025 state budget, though many are more 'catch-up' growth area additions than established area network reform as envisaged in the Bus Plan. 

Conclusion 

The audit's findings are a matter for the Auditor-General after weighing all evidence.

However based on what is known publicly I would imagine its conclusions may be more than a 'good job keep it up' type result. After all VAGO performance audits almost always find something that can be improved, even if minor. 

Consequently the Bus Plan audit promises to be interesting with meaty recommendations that could give the department some worthwhile guidance on improved performance. Especially if laced with big helpings of reality from auditors who are not so remote as to forget to occasionally GOTB! 

Regional Rail Link turns 10 / Metro Tunnel testing

Today is a super important date in transport for two reasons. One related to the past, the other for the future. 

Regional Rail Link & reformed buses

Firstly regional Rail Link turns 10 today. This transformed transport in Melbourne's west and Geelong with the new Regional Rail Link starting, bringing rail to new stations at Wyndham Vale and Tarneit. We had built new stations and extended electrification but this was a completely new line on a completely new alignment.

While the intent was to provide a bypass to free up rail capacity in western Melbourne, the new stations became massively busy in their own right due to them serving huge growth area catchments popular with migrants seeking new affordable housing. Tarneit, for example, is now V/Line's busiest station outside Southern Cross. This 2010 Paul Mees ATRF paper, throwing shade on the Regional Rail Link, has not aged well given the undoubted patronage success of its stations; it is now impossible to see western Melbourne without it. 

The basic off-peak weekday service to South Geelong started as every 20 minutes, versus the hourly provided under the old alignment via Werribee. Patronage boomed with this link becoming crowded shortly after opening. Weekend service started at hourly on opening. However crowding forced that to every 40 minutes. That wasn't enough so some extra trips were slotted in to provide some 20 minute intervals. Finally last December a 20 minute 7am - 9pm weekend timetable was instituted with more trips extended to Marshall. 

The Regional Rail Link has one 'good' problem -  it has become a victim of its own success. There is massive demand for more than the original two stations. There's been talk of extra stations but none have opened in RRL's first decade. Although West Tarneit is under construction with an opening due next year.

Reluctance to add new stations may be due to a view that this will crowd trains and unacceptably slow travel for Geelong passengers. RRL's future is clearly a two-tier service, with an intensive (possibly electric) service to Wyndham Vale and express services to Geelong/Marshall/Warrnambool. There also needs to be a westward extension of Werribee Metro electrification with a new station at Black Forest Road.  

Wider network reforms - varying prospects

Regional Rail Link in 2015 was not just a V/Line rail upgrade project. As originally conceived it was also going to be the centrepiece of massive multimodal Metro train and bus network revamps across Melbourne. Planning for this had been done but their fates varied.

While the new Labor government pursued its infrastructure program with gusto, it resisted adding Metro service, especially if new timetables would affect travel for some in (then) marginal seats on the Frankston line. Thus the proposed 2015 Metro timetable was ditched, setting rail service reform back years, with elements not picked up until 2021 (and again likely late 2025). This delay meant that some anticipated benefits of RRL, that of freeing space for more Werribee line Metro services, were not realised as soon as they could have been, as the Auditor-General noted in 2018. A decade on we're still waiting for some, though last month's state  budget has funded some welcome Werribee peak uplifts.   

The government wasn't just shilly-shallying on Metro rail service reform. The proposed 2015 Transdev greenfield bus network was also scrapped . Although some reasons for not proceeding with some of this were sound, a promised later review did not happen and the cause of bus reform was put back years.  

Fortunately the radically reformed RRL-related bus networks designed for Geelong and Wyndham survived. Geelong's new 'greenfields' network featured simpler more direct routes running at higher frequencies. Apart from isolated and minor objections this simplified Geelong bus network has proved successful with operating hours its main outstanding concern. 

RRL could have opened without the Geelong bus changes. But the bus revamp enabled PTV to present a compelling multimode package to the public on Day One. This is a recipe Perth's Metronet routinely follows with new bus networks for Airport, Yanchep, Ellenbrook and Thornlie-Cockburn. Melbourne's record here is only sporadic; if we were as good as Perth now (or our own record with Geelong in 2015) our Metro Tunnel would get a radically revamped bus network between at least Watergardens and Dandenong along these lines along with reformed CBD trams. The indications so far are not strong for sweeping changes here, although some smaller bus reforms around Parkville happened last year. 

Wyndham also got a brand new bus network on this day in 2015. In contrast to Geelong that had to happen as otherwise there would have been few if any routes properly serving the new stations at Tarneit and Wyndham Vale. That would have caused massive parking pressures and jeopardised public goodwill towards the RRL.

The 2015 Wyndham bus network comprises a two-tier network with more frequent main road routes and neighbourhood style coverage routes. While service levels aren't yet as high as they need to be the network has been extremely successful with Wyndham bus routes ranking amongst the most productive in Melbourne. This has set up Wyndham buses on a path of continuous improvement, as opposed to the network atrophy seen in other areas (eg Melton town, Broadmeadows, Preston - Epping, Ringwood, Knox, Greater Dandenong, Frankston etc). The latest Wyndham bus boost will start in less than two weeks, with improved timetable for four routes starting on July 1.  

See my 4th anniversary write-up here for more about the Regional Rail Link. 

Metro Tunnel's full day test

This one's about the future. 

Rail officials and enthusiasts alike are out and about all day today to watch a full day's testing of the Metro Tunnel timetable. 

While it's a Saturday, Rail Express reports that a weekday timetable will be operating on lines that will use the Metro Tunnel. 

Passengers will need to change at Caulfield or Footscray if travelling through to the city.

This won't matter to observers as they can just find a seat somewhere, grab their watch and tick off train arrival times. While this is normally only an activity undertaken by signalling staff and hard-core gunzels, this time it's of much wider interest. 

This interest is partly because the state government has kept even basic Metro Tunnel frequency information under wraps. 

Despite adding relatively little Metro train service in its first decade, it has raised expectations of the service possible with the Metro Tunnel and other projects. This heightened anticipation has given rise to a lot of guessing and discussion including here.

Speculation on Metro Tunnel service levels intensified last week when PTV's website showed modified timetables for Cranbourne, E Pakenham and Sunbury, applicable for today, that could well be part of a Metro Tunnel weekday timetable.

Daniel Bowen analysed this here . He hoped it was only partial, with the real timetable having more trips. Such a consistently high all-day frequency is key to the sort of multi-directional changing all day activity that is needed to make the Metro Tunnel a patronage success.  

It is also desirable that best endeavours are made for the final timetable to avoid big 'holes' such as the peak period / peak direction 20 minute gap towards Cranbourne shown below.  

The 2012 Network Development Plan - Metropolitan Rail proposed a hierarchy of base all-week frequencies for the rail network, including a 5 min inner core interval. Peak frequencies would be higher. For example it was envisaged that Cranbourne, Pakenham and Sunbury (all Tier 2 stations) would have 5, 9 and 9 trains per hour respectively by 2022 (see Figure 5-8). 

The pandemic and its aftermath have reduced peak period train usage to less than envisaged in the NDP. However a 5 minute all day core frequency between at least Footscray and Caulfield remains essential for the Metro Tunnel to provide a 'big city' metro user experience, justify its construction costs and play its full role in the central area transport network (including enabling cascading tram network reform).   

It will be interesting whether observations people make today tally with published timetables or not. And if any media releases reporting on the success or otherwise of today's trials provide further information on Metro Tunnel service levels.

Any observations on today would be welcome and can be left in the comments below. 

UN 204: Kingston proposes better buses for the south-east

 

The City of Kingston is advocating for better bus services in Melbourne's south-east. It has set up a website listing three major and three longer term advocacy priorities. After public consultation on them finishes next month these will be submitted to Department of Transport and Planning in September. There has already been media coverage of Kingston's bus advocacy including in Australasian Bus & Coach and Dandenong Star Journal. 

This advocacy comes at a desirable time. The 2025 state budget marked a revived interest in bus services though all but one of the new initiatives were in Melbourne's west and north. However 2026, being the last budget before the state election, offers an opportunity for advocates (including councils) to be bolder in what they ask for. With the potential for what doesn't get budget funding to be picked up by one or more of the political parties in the campaign for November's state election. 

Kingston's six route wish-list is mapped below. 

Key and major priorities

Key priority is the east-west Route 828 between Hampton and Berwick via Southland and Cheltenham. This is a long route with a lot of residential catchment and destinations away from the rail network. Not just in the City of Kingston but also in Bayside, Greater Dandenong and Casey. While it runs every 20 minutes on weekdays its weekend service falls off to every 40 minutes on Saturdays and 60 minutes on Sundays. Kingston wants this upgraded to run every 20 minutes every day, working the existing bus fleet harder. 

Route 708 and 903 are described as major priorities. Route 708, operating every 30 min on weekdays and 60 min on weekends, serves nearly all of southern Kingston away from the Frankston line. It also passes near the under-construction Mordialloc Aquatic Centre. Council is seeking a weekend upgrade to every 30 minutes. It would also like to hear views on whether the 708 should be deviated via Aspendale Gardens Shopping Centre, something that would be convenient for some but would slow other peoples' trips. 

Route 903 is Melbourne's busiest single bus route with many major destinations en route. Operating from Mordialloc to Altona it is one of our three orbital SmartBuses. Weekday service is every 15 minutes but weekend service has 30 minute gaps - not ideal for a premium route. Kingston wishes to see the busy Mentone - Box Hill section of this route improved to operate every 10 minutes on weekdays and every 15 minutes on weekends.  

These three service upgrades would improve public transport connectivity across a large section of Kingston, particularly on weekends. The main expense would be extra bus driver hours. Routes 828 and 903 (especially) are above average patronage productivity with improved service likely to stimulate further usage growth (as happened when Route 800 on Princes Hwy gained greatly improved weekend service last November). I suggested all three upgrades as bus advocacy priorities for Kingston last year. 

Longer term priorities

The above major priorities are relatively simple to implement as they are upgrades of existing routes. Improvements can happen as little as six weeks after budget funding if there is sufficient political will as seen with Werribee's bus upgrades starting soon.

Adding new routes or reforming existing routes tends to take longer. But this is still important to tackle network gaps or inefficiencies. Kingston has three longer term priorities here, as follows: 

Bay Road corridor bus. Currently there is no direct public transport between Southland Shopping Centre and Sandringham station despite there being a direct road there. Neither does Southland station have bus routes nearby. The closest alternatives are routes 708, 822 and 828 which are both indirect and not consistently frequent. A Bay Rd connection could be formed with a new route or (more economically) rerouting an existing route like the 828 direct from Southland to Sandringham. I have written about the need for a Bay Rd bus many times including in my list of bus advocacy priorities for Bayside. 

Southland - Elsternwick Nepean Hwy bus. Nepean Hwy is a major thoroughfare but significant sections have no bus. Other parts do have a bus but it is the occasional Route 823 from Southland to Brighton once an hour on weekdays only. A reappraisal of the local bus network would likely include a Southland - Moorabbin - Elsternwick bus that would fill this 'missing gap' in the network. Again this route features in the bus advocacy priorities for Kingston I prepared last year. 

Mordialloc - Clayton - Monash University bus. Current public transport access from southern Kingston to the Monash employment and education precinct is very poor. The state government has spent hundreds of millions building the Mordialloc Freeway for these type of trips but has done zero for public transport on this corridor. Access to local jobs is also a problem with Braeside being a public transport desert save for a handful of Route 705 trips a day from Mordialloc (which instead of continuing north to Clayton turns off to Springvale). A new Mordialloc to Monash University bus would be more expensive than some of the other upgrades discussed here but would make many trips faster and easier than now. I investigated the possibility of such a route back in 2019 here.  

What else could have been advocated?

I'd have liked to have seen a Route 733 extension to Southland, Cheltenham or potentially even Sandringham along the lines of the Suburban Rail Loop SmartBus concept. While appealing, the implementation would be slower than straight service upgrades due to the need to tie in with bus network reform in the Oakleigh/Clarinda area.

Other potential inclusions include improved weekend frequencies and simplification of the 811/812, improved Route 824 weekend frequencies in Clarinda and 7 day service for the very infrequent 857  between Chelsea and Dandenong. Better weekend service on the crowded 902 orbital is arguably even more important, though much of it runs on the Springvale Rd border between Kingston and Greater Dandenong, with its busiest portions being in Greater Dandenong and Monash more than Kingston. 

A risk that may have ran through Kingston's mind is that asking for too much can complicate messaging and mean that nothing requested happens. If you accept that then the six options chosen is probably a good balance. 

Conclusion

The City of Kingston has come up with six proposals that if adopted will greatly improve bus travel not only in the City of Kingston but also adjoining municipalities including Bayside, Monash, Box Hill and Greater Dandenong. 

The real test is how the state government reacts. While DTP might back the proposals they would all require budget funding that requires political support.

It is for this reason that I urge people that as well as completing the Kingston bus survey they request the support of state parliamentarians to advocate for. fund and deliver these essential bus service upgrades. 

See other Useful Network items here

TT 206: Werribee's off-peak bus upgrades done pronto

The state government has set a new standard for upgraded bus services in Melbourne's outer suburbs. Not only in frequency and operating hours but also in a radically compressed budget funding to delivery lead time. 

With DTP typically needing 12 months to deliver even simple bus timetable upgrades after budget funding, observers thought that the 7 months achieved for last year's Route 800 upgrade was quick. But now the department has shrunk budget-implementation lead time to just six weeks for this very worthwhile package that adds 470 extra trips per week. The difference a tight by-election, political will and/or new delivery-oriented DTP leadership can make! 

The upgrades, funded in last month's state budget, will see later trips and/or improved frequency on bus routes 170, 180, 190 and 192 from Werribee Station. They will take effect on Tuesday July 1, the first day of operation under new bus franchise agreements (though there's no operator change with CDC  continuing to run Werribee buses).

Operating hours

Melbourne bus routes typically start later in the morning and stop about 3 hours earlier than trains with 9 to 9:30pm finishes common on most nights. 

These upgrades extend service until about midnight with smaller extensions (to approx 10pm) on Sunday nights. This means that at nearly all times people will want to catch a bus - day or night - there will be a service on four key bus routes in the City of Wyndham. That's three routes up from now where only the Route 190 currently operates long hours. There is however a catch affecting one route - more on that later. 

Frequency

Buses are also usually less frequent than trains. These improvements help here too. Currently these routes drop to every 40 minutes after about 7 or 8pm on weekdays and 5 or 6 pm on weekends. 

Next month's upgrades add trips such that the service remains at every 20 min until approximately 10pm. Then it falls off to every 40 minutes. This expansion of daytime frequency to about 9 or 10 pm on all nights of the week for routes 170, 180 and 192 matches improved Geelong train timetable last year, Route 905 and 907 bus upgrades this year and potentially also the Metro Tunnel train timetable later this year. Route 190 also gets some improvements but (contrary to confusing PTV wording) weekend service remains at every 40 minutes. 

Out of the four routes the 192 is the biggest winner with both longer hours and doubled 7 day frequency. As well as covering a large unique growth catchment the government will be hoping that this bus consoles people who really want the widely expected station at Black Forest Rd. However the bus won't be a satisfactory option for commuters until its infrequent peak timetable is fixed - something that, in the rush to deliver some improvements as soon as possible, was out of scope for this upgrade.    

Have you read about any of these upgrades before? 

Possibly. What is proposed is very similar to some of these upgrades discussed in 2019 and 2024. 

Potential future improvements

These are big improvements but there's always others that could make Wyndham area buses even more useful. Here's a few: 

* Fixing low peak frequencies on all four routes: Excluded from these timetable changes is peak frequency. All four routes retain gaps of around 30 and even 40 minutes in the morning peak. The timetable for Route 180, for example, has a 28 minute gap at the peak of the peak arriving at Werribee (there being no arrivals between 7:21 and 7:49am). Parts of 180 at least have the 182 overlapping on Tarneit Rd. But this cannot be said for the 192 which remains with 30 to 35 min peak gaps. 190's gaps are almost as long while the 170 has a 41 minute peak period gap.

The Metro Tunnel will see a rewrite of the Werribee line train timetable including a 7.5 minute peak frequency. Hopefully these are stage 1 upgrades with more coming after more buses arrive. It would be highly desirable for main bus routes to be upgraded to operate every 15 minutes in the peaks, noting that this will increase the peak bus requirement.

* 192 - fixing late weekend morning starts: Despite long operating hours in the evening the weekend morning start times are unacceptably late, especially for a bus route with a lot of unique catchment. For example the first arrival at Werribee is 8:20am Saturday and 8:38am Sunday (much like the current timetable). These need to be made 60 - 90 min earlier, even if it means some early morning and evening gaps are widened to 40 minutes to keep the number of weekly trips the same. 

* 190 - weekend frequency: PTV's wording does not match the timetable here with this exacerbated by the same wording being used for a different service pattern and customary idiom (see box below). Clarity is essential when you have many users whose first language is not English. Contrary to what PTV says, Route 190 will remain at every 40 minutes on weekends rather than joining 170, 180 and 192 with 20 minute weekend frequencies. But getting the 190 up to every 20 min all week would be a good idea, increasing the proportion of people within walking distance of a 7 day 20 minute service. It would also remove the network oddity of Route 190 having wide operating hours but a low weekend frequency.  

* 7 day 20 minute service on other routes in Wyndham. These upgrades are good but open up an inequality where there remain strongly used routes in the east of the municipality that still have 40 minute gaps. Popular Wyndham routes that deserve a 7 day upgrade to every 20 minutes include the likes of 150, 152, 160, 182, 494, 495 and 497. 

Conclusion

These are a good set of off-peak bus timetable upgrades for an area whose residents will definitely use them. Others for Wyndham, including the 153 upgrade and new routes will come later. Closing the embarrassingly long peak period gaps for the above four routes to 15-20 minutes should also be high priorities as soon as new buses arrive. 

Index to other Timetable Tuesday items

Which MP is Victoria's transport question king?

Which state politician asks the most questions about transport in parliament? 

Keep reading and you'll find out who is Victoria's transport question king. 

At their best, questions asked in parliament are meant to hold the government to account. 

At other times they may be theatre. They may be intended to embarrass more than seek elucidation if coming from the non-government side. Or if from the government side they may be sympathetic 'Dorothy Dixers' used to highlight some good news, announce some funding or parade some achievement.   

The Victorian Parliament website has a searchable questions section. There you can see how many questions each MP has asked on which portfolios. 

I decided to do this exercise for the 60th parliament - that is the time from after the 2022 state election to now (4 June 2025 when this data was collected). I counted only questions asked for the transport portfolio (1988 in total). That needed six boxes ticked due to machinery of government changes introduced part way through the term. 

Counting questions asked in parliament is just one measure of how active MPs are and/or their interest in transport matters. Note the emphasis. Different MPs have different representation styles. Some, especially on the government side, prefer to press their area's needs through personal meetings with the minister. There may also be liaison at electorate officer/ministerial adviser levels to refine proposals. 

Much of a diligent MP's time is away from parliament, often in their electorate, showing up to things and talking to constituents. An MP that only asks a few questions is not necessarily lazy; indeed some work hard and can claim significant wins for their seat through more effective means. Indolent or self-serving MPs do exist but it is unfair to use questions asked as a sole gauge of this. 

Having got these caveats out of the way, here are the lists of MPs by house in descending order of questions asked. As Question Time is supposed to be a means for the parliament to hold the executive to account, all those asking questions are non-ministers. Where MPs have left or been expelled from parties, I've listed the party they represented at the 2022 election first. Government and non-government parliamentarians are separated as their need and propensity to ask questions is different, especially in the Legislative Council. 

LEGISLATIVE ASSEMBLY

Government

Will Fowles (ALP/Ind, MLA) 25

Sarah Connolly (ALP, MLA) 18

Luba Grigorovitch (ALP, MLA) 16

Kathleen Matthews-Ward (ALP, MLA) 16

Nathan Lambert (ALP, MLA) 14

Anthony Ciaflone (ALP, MLA) 13

Paul Hamer (ALP, MLA) 10

Juliana Addison (ALP, MLA) 10

Mathew Hilakari (ALP, MLA) 10

Jackson Taylor (ALP, MLA) 10

John Mullahy (ALP, MLA) 9

Iwan Walters (ALP, MLA) 9

Alison Marchant (ALP, MLA) 8

Dylan Wight (ALP, MLA) 7

Tim Richardson (ALP, MLA) 7

Kat Theophanous (ALP, MLA) 7

Martha Haylett (ALP, MLA) 7

Daniela De Martino (ALP, MLA) 7

Bronwyn Halfpenny (ALP, MLA) 6
Lauren Kathage (ALP, MLA) 6

Josh Bull (ALP, MLA) 5

Gary Maas (ALP, MLA) 5
Emma Vulin (ALP, MLA) 5

Paul Mercurio (ALP, MLA) 5

Jordan Crugnale (ALP, MLA) 5

Nina Taylor (ALP, MLA) 4
Matt Fregon (ALP, MLA) 4

Katie Hall (ALP, MLA) 4

Michaela Settle (ALP, MLA) 4

Ella George (ALP, MLA) 4

Meng Heang Tak (ALP, MLA) 3

John Lister (ALP, MLA)* 2

Eden Foster (ALP, MLA)* 1

Pauline Richards (ALP, MLA) 1
Steve McGhie (ALP, MLA) 1

Paul Edbrooke (ALP, MLA) 1

Darren Cheeseman (ALP/Ind, MLA) 1

Nick Staikos (ALP, MLA) 1

Vicki Ward (ALP, MLA) 0

Chris Couzens (ALP, MLA) 0

Non-Government

Annabelle Cleeland (Nat, MLA) 81

Nicole Werner (Lib, MLA) 58

Danny O'Brien (Nat, MLA) 55

Tim Read (Grn, MLA) 52

Cindy McLeish (Lib, MLA) 52

Tim Bull (Nat, MLA) 43

Ellen Sandell (Grn, MLA) 41

Bridget Vallence (Lib, MLA) 32

Kim O'Keeffe (Nat, MLA) 30

Kim Wells (Lib, MLA) 30

Wayne Farnham (Lib, MLA) 29

Brad Rowswell (Lib, MLA) 25

Martin Cameron (Nat, MLA) 21

David Southwick (Lib, MLA) 19

Bill Tilley (Lib, MLA) 19

Tim McCurdy (Nat, MLA) 18

Roma Britnell (Lib, MLA) 15

David Hodgett (Lib, MLA) 14

Richard Riordan (Lib, MLA) 14

Sam Hibbins (Grn/Ind, MLA)* 13

Jade Benham (Nat, MLA) 13

Sam Groth (Lib, MLA) 11

Brad Battin (Lib, MLA) 9

James Newbury (Lib, MLA) 9

Matthew Guy (Lib, MLA) 7

Michael O'Brien (Lib, MLA) 7

Peter Walsh (Nat, MLA) 7

Emma Kealy (Nat, MLA) 4

de Vietri (Grn, MLA) 3
Chris Crewther (Lib, MLA) 3
John Pesutto (Lib, MLA) 3

Rachel Westaway (Lib, MLA)* 1

Ryan Smith (Lib, MLA)* 0

(*) Member has not been present for full period of the 60th parliament (due to resignations or by-election wins). 

As you might expect the most persistent questioners did not sit on the government benches. Possibly less expected was that members of smaller parties like Nationals or Greens punched above their weight relative to many Liberals. Not surprisingly, given the areas they represent, maintenance of country roads was a top concern for National MPs. 

On the government side the most active questioners were metropolitan members from the west and along the Ringwood line (the latter of which has been marginal for Labor). Quieter Labor MPs (at least on transport) are found around Geelong and parts of the south-east. 

You would expect that non-government members who hold their parties' transport portfolios would be particularly active questioners on the topic. The numbers do not show that. 

David Southwick (who had the transport infrastructure role until January 2025) asked 19 questions. Matthew Guy (who had public transport since October 2023) asked 7 questions. The Greens Sam Hibbins (who had his party's transport spokesperson role until claimed by scandal) was more active but still below average with 13 questions asked.

Previous shadow transport ministers in the Legislative Assembly (David Hodgett and Richard Riordan), each with 14 questions, do not stand out although one MLC does as mentioned later. Transport was also rarely top of mind for current Liberal leader Brad Battin (9 questions) and former leader John Pesutto (3 questions).  

LEGISLATIVE COUNCIL


Government

Michael Galea (ALP, MLC) 19

Ryan Batchelor (ALP, MLC) 8

John Berger (ALP, MLC) 6

Sheena Watt (ALP, MLC) 5

Tom McIntosh (ALP, MLC) 4

Sonja Terpstra (ALP, MLC) 3

Jacinta Ermacora (ALP, MLC) 2

Lee Tarlamis (ALP, MLC) 0

Non-Government

David Davis (Lib, MLC) 194

Evan Mulholland (Lib, MLC) 100

Wendy Lovell (Lib, MLC) 90

Richard Welch (Lib, MLC)* 52

Aiv Puglielli (Grn, MLC) 41

Trung Luu (Lib, MLC) 39

David Ettershank (Can, MLC) 35

Rikkie-Lee Tyrrell (PHON, MLC) 35

Katherine Copsey (Grn, MLC) 32

Joe McCracken (Lib, MLC) 32

Nick McGowan (Lib, MLC) 31

Bev McArthur (Lib, MLC) 26

Gaelle Broad (Nat, MLC) 23

Ann-Marie Hermans (Lib, MLC) 21

Renee Heath (Lib, MLC) 20

David Limbrick (LP, MLC) 20

Moira Deeming (Lib, MLC) 19

Melina Bath (Nat, MLC) 18

Samantha Ratnam (Grn, MLC)* 17

Sarah Mansfield (Greens, MLC) 13

Rachel Payne (Can, MLC) 13

Matthew Bach (Lib, MLC)* 9

Georgie Purcell (AJP, MLC) 9

Georgie Crozier (Lib, MLC) 7

Anosina Gray-Barberio (Grn, MLC)* 4

Adem Somyurek (Ind, MLC) 3

Jeff Bourman (SFF, MLC) 2

(*) Member has not been present for full period of the 60th parliament (due to resignations or Legislative Council appointments). 

The Legislative Council is quite different to the Legislative Assembly. It is explicitly meant to be a 'house of review'. And it has fewer members, meaning that members who choose to get to ask more questions. It may also help that the government lacks a majority here. 

Most notable (unlike in the Legislative Assembly) is the strength of the Liberals. Most notably David Davis, Evan Mulholland, Wendy Lovell and Richard Welch. All asked over 50 questions with the Suburban Rail Loop a popular topic. Greens, Legalise Cannabis and One Nation MLCs feature in those who asked over 30 questions.

Unlike in the Legislative Assembly those who were or are transport spokespeople for their party (David Davis and Katherine Copsey) were active question askers. 

With few exceptions, Labor Legislative Councillors rarely ask questions, unlike their counterparts in the Legislative Assembly. Of the Labor MLCs Michael Galea stands out, with most others asking five questions or fewer. 

Conclusion

MPs vary greatly in their tendency to ask questions in parliament. 

As might be expected non-government MPs ask more than government MPs. 

There are more questions asked in the Legislative Council than the Legislative Assembly.

Liberals dominate questioning in the Legislative Council whereas the biggest questioners in the Legislative Assembly are Nationals and Greens. 

Labor Legislative Councillors rarely ask questions whereas Labor MLAs do more often. Possibly as MLCs are effectively party appointees who rarely get large personal votes. Whereas MLAs are expected to (within limits) represent their single-member seat, including sometimes asking questions.  

The opposition tends to be the reverse; Liberal MLAs under-ask relative to their counterparts in the Legislative Council. 

It is one of the latter, David Davis MP, who is unmasked as Victoria's transport question king by asking 194 questions so far this parliament. 

Perhaps more MPs, especially Liberals in the lower house and Laborites in the upper house, should have more of what he's drinking. 

TT 205: More Trains More Often - How is the government going & Metro Tunnel prospects

Last week and this has been the budget estimates hearings. The time when ministers explain and are asked questions on the recent state budget and the government's record generally. Sometimes additional information can come out or fine points get clarified. 

For example last Thursday we learned that the Metro Tunnel will open with all stations operating (there was some media speculation that this wouldn't be so) and that the 463 bus would be extended to Caroline Springs. Watchers also heard the customer experience jargon term 'hypercare' in relation to getting people used to the Metro Tunnel. 

Also of interest was the government defending its record on Metro train service frequency. In a nutshell it hasn't been good, with falling service per capita. The big picture is this: 

Melbourne's population grew 18% since 2015
Metro train service only grew 9% since 2015

Every time people have asked for shorter gaps between trains (like Sydney or Perth already have) the government has said that this or that infrastructure project must be done first. Or that it will all be amazing after the Metro Tunnel opens - people just need to be patient. 

But there's only so long that the government can string people along, and ten years is more than enough. Endless line shutdowns and few if any service improvements at the end of them is wearing thin with passengers. Even the cheapest timetable upgrades that would halve the longest waits on key lines got pushed into the never-never as government interest in Metro frequency improvements collapsed after 2015-2016.   

Anyway that's the narrative you've read about here and elsewhere. 

More trains more often - the record

What does the government say about this?  

Last Thursday the minister showed a slide depicting its record of train service additions over the last decade. These are shown in the small circles (blue for Metro, purple for V/Line). The larger circles represent initiatives funded in last month's budget (to be implemented).

The numbers are weekly services added. That can make even small service additions look big. For example adding one return trip per day on one line is 14 extra trips per week. Multiply by 15 lines to get a massive 210 trips per week. 

That's a handy trick to make it look a big deal. Until you realise that because Metro runs over 15000 trips per week, that's just a 1.4% uplift in service. Helpful but not revolutionary.   

The asterisked note at the bottom left is hard to read but says that 50 of the 200 weekend services funded in the 2022-23 budget are still in delivery and are not shown. These include additional weekend V/Line services for Seymour, Shepparton, Traralgon and Bairnsdale. 

You could try clicking on the screenshot below but it may still be unclear. So I'll go through it below, starting with the ten year record. 

2015 - 2024 service additions

Below I'll explain what the Metro network got over the last decade in terms of service. 

First let's look at what the whole network got thanks to Night Network in 2016. This added hourly weekend trips on all lines between about 1am and 5am Saturdays and 1am to 7am Sundays. That's roughly 24 trips per week added on every line (about 6 each way multiplied by 2 for return, multiplied by 2 again for each day).

Thus if a line has about this number of trips added between 2015 and 2024 you can attribute its entire increase to this one Night Network initiative. This applies on the Belgrave (+18), Alamein (+23) and possibly Glen Waverley (+30) lines. You could say that leaving aside Night Network service on these lines has been stagnant over the last decade. 

Now to other lines. Sunbury line got 68 extra trains per week. Over 40 of these would be non-Night Network. Some could have been evening extensions circa 2016 (previous to that about half the evening trains terminated at Watergardens, giving Sunbury only an hourly service) with the others being some extra peak or shoulder peak runs. The best is still to come for Sunbury with this line likely to be the biggest proportional beneficiary from the new Metro Tunnel timetable. 

Also awaiting the Metro Tunnel are the Pakenham and Cranbourne lines. These got a combined 201 extra trips per week. Subtracting Night Network drops this to about 150 trips per week extra. Examples of additions include (i) just after the opening of Skyrail in 2018 (more weeknight trips), (ii) some further boosts in 2020 and (iii) 50 more weekly services when the Cranbourne duplication got done in 2022. None of these changed maximum waits (30 min weekend evenings and up to 70 min Sunday mornings) but expectations are high that the Metro Tunnel timetable starting later this year will shorten these.  

Upfield can claim 57 extra trips over the decade to 2024. Subtracting Night Network that's about 30 trips per week or 3 each way per weekday. These contributed to some minor peak frequency upgrades (ie from every 20 to every 15 min approximately). 

Werribee / Williamstown / Frankston - meant to get a boost in 2015 to coincide with RRL but didn't with some argy-bargy over Frankston trains running direct or via the loop (or maybe half and half - which was tried and proved a mess).

It stalling on the Network Development Plan (Metropolitan Rail) was the single biggest reason for this government's record on Metro train frequencies being so undistinguished to date. Anyway things finally got tidied up in 2021 with evening and Sunday morning maximum waits cut to 20 minutes on all three lines and peak services for Altona and Williamstown restored from 22 to a simpler 20 minute headway. This and other boosts gave Werribee 186 extra weekly trips, Williamstown 119 extra and Frankston 117 more. 

After the 2025-budgeted Upfield and Craigieburn boosts happen the Mernda and Hurstbridge lines will stand out as having the longest waits outside the peaks, particularly on Sunday mornings. This is notwithstanding previous gains for Mernda (175) and Hurstbridge (75 weekly trips). This is because, Night Network excepted, all these increases were on weekdays, mostly peak and shoulder peak, after duplication works in 2023. The government can claim a major achievement in 2018's Mernda electrification, but it missed the opportunity then to add the relatively few weekly trips needed to cut maximum waits on the rest of the line. 

The Burnley group has had a lot of level crossing removals such that the Lilydale line is now level crossing free. This included a consolidation of stations with Surrey Hills and Mont Albert combined into the new Union station. Peak timetables remain complex with numerous stopping patterns and interpeak gaps beyond Ringwood remain at 30 minutes - longer than two regional lines. Box Hill has built up but its train timetable has remained basically stagnant, inferior at night to even Fawkner Cemetery once the Upfield line gets its evening frequency boost. In the decade from 2015 Lilydale gained 45 services per week, or approximately 2 each way trips per weekday after the Night Network trips are excluded. It's not a lot, especially as the Burnley group now has a lot of marginal seats along it and Box Hill is now densely populated and high-rise. 

Train services versus population growth 

Below is all the implemented Metro service gains put on a table. They add to 1252 per week. Is this a big or a small service uplift, noting that we are talking about ten years worth? To know we need to know how many services Metro run to arrive at a percentage increase. Metro currently claim 'over 15000 weekly services' on their Linked-In. 

Comparing the 1252 trips with an assumed 14000 base in 2015 we arrive at a 9% increase across the Metro network.  In contrast population grew by 18% over that time. Thus we can say that Metro service provision has grown at only half the rate of population. In other words a significant fall in service per capita and a lag behind cities such as Sydney and Perth whose stations typically enjoy more hours of frequent service per week.  

2025's funded upgrades (and the Metro Tunnel) 

Signs that this government may finally be taking service frequency seriously appeared in last month's state budget. As you can read here this funded a decent package of Metro train and bus service boosts in Melbourne's north and west plus an interpeak boost on the Sandringham line. 

The annotated map tabled in PAEC listed gains for Werribee, Craigieburn, Upfield and Sandringham lines. As already publicised, Werribee will gain higher peak frequencies with 20 trips added. 20 trips per 5 day week is 4 per day. Two trips per hour extra are needed to boost peak frequency from 6 to 8 trains per hour (ie the 10 to 7.5 min frequency improvement promised). That 4 extra trips per day means the boosted peak service applies for one hour in both directions or two hours in the peak direction only (if train numbers and stabling permits this). 

Craigieburn and Upfield are shown with 100 trips added to deliver a 20 min maximum wait. However I'm not sure about the workings - just to get rid of the 30-40 min evening and Sunday morning gaps would likely require 70 more trips per week on each line. And that doesn't count the shoulder peak uplifts that the Craigieburn line will be getting. So the actual increase may be more than indicated. Unless something creative is done like spreading some peak trips into the shoulder peaks and evening. 

The Sandringham line gets an interpeak weekday uplift from 4 to 6 trains per hour. The number of increased trips is not readable. But you can work out that it's 4 trips extra per hour (accounting for both directions) over a 6 hour span and 5 day week then it's in the region of 120 trips per week added. 

The above adds to roughly 250 weekly trips funded in the budget on lines other than the Metro Tunnel. That is just under a 2% increase on Metro's current 15 000 weekly trips.

That 250 trips leaves about 1000 more weekly trips needed for Melbourne to return to the per capita Metro service that this government inherited when it won office in 2014. 

When it faces the people in November 2026 will this government even be able to claim that they maintained per capita Metro train service? With the per capita decline to date, the answer depends heavily on the service rebound the Metro Tunnel timetable will deliver later this year.   

That wasn't on the minister's budget estimates presentation slide. Hence the absence of big blobs near Sunbury and Dandenong. But they will likely exceed the ~250 increase. Some hypothetical back-of-envelope examples:  

BASIC METRO TUNNEL SERVICE
(Watergardens - Dandenong every 10 min day/20 min night with higher peak frequency)

* Boost Sunbury line interpeak weekday services from every 20 to every 10 min: ~200 extra trips per week (+3 trains per hour over 6 hours x2 for each way x 5 days of the week)  

* Double Sunbury line weekend services: ~200 trips (based on doubling current service of approx 50 trips each way each weekend day).  

* Boost Sunbury line weeknight services from every 30 min to every 20 min: ~40 trips (+1 train per hour over 4 hours x 2 for each way x 5 days of the week) 

* Boost Cranbourne & Pakenham weekend evening & Sun am from every 30 to every 20 min: ~60 trips (+1 train per hour over 5 hours x 2 for each way x 2 nights of the week x2 lines & Sunday am uplift)

Total ~500 extra trips per week

Any Metro Tunnel timetable that does not deliver at least the Basic service above will be laughed at. The government should be smart enough to know this given its political and financial investment in the project over its entire decade in office. Demonstrated success with the Metro Tunnel switch-on will also boost its credibility with regards to other transport projects, particularly the Suburban Rail Loop, which it needs right now. Furthermore, in addition to the ~250 weekly Metro services on other lines that this year's budget funds, the 750 trip total will deliver more then half the service needed to return to 2014's  Metro service per capita number. 

Want the Metro Tunnel to be world-class? Or even Sydney class? It wouldn't be a great reward for taking some of the southern hemisphere's longest escalators if you've got another 9 (or 19) minutes to the next train if something like the basic minimum above was implemented. We also wouldn't compare well with the 5 minute service on Sydney's Metro or even the 7.5 minute interval common at inner Perth stations. 

Justifying the Metro Tunnel's construction costs, maximising land use and development synergies and unlocking the wider CBD benefits of complementary tram network reform all require something better with high all week frequencies like we've never seen before. Potential ingredients of such an enhanced timetable (and the approximate number of weekly trips added) are below. 

ENHANCED METRO TUNNEL SERVICE 
(Watergardens - Dandenong every 5 min day/10 min night with higher peak frequency)

All the above plus: 

* Boost Watergardens - Dandenong weeknight service to every 10 min to late: ~200 trips (+3 trains per hour over 3 hours x 2 each way x 5 days per week x 2 lines) 

* Running Metro Tunnel central section every 5 min 10am-4pm weekdays: ~360 trips (+6 trains per hour over 6 hours x2 each way x 5 days of week)

* Running Metro Tunnel central section every 5 min 9am - 9pm weekends: ~300 trips (+6 trains per hour over 12 hours x2 each way x 2 days of week) 

* Peak upgrades as needed to address crowding (not counted) 

* Upgrades and reforms to connecting trains, trams and buses to improve network effects (not counted)  

Total ~ 1400 extra trips per week (including 500 from basic option above)


The 1400 extra trips plus the 250 on other lines is a total of 1750 extra trips.

If implemented this would mark a significant acceleration of Metro service uplift given that the government only added 1250 trips in its first decade.

On a 2015 base of 14000 trips, this uplift of 3000 to 17000 trips would be a rise of 21%. That would slightly exceed the 18% population growth since then. Then the government can claim to have increased per capita Metro services, which it cannot at the moment.

Regardless of the service we get from the Metro Tunnel, there will still be big service inequalities between lines. That includes lines that are (a) busy, (b) serve growing or densifying areas, (c) serve populations with high social needs and (d) all three. That makes further frequency uplifts desirable so that you can rock up at any station (within the urban growth area) and not have up to 40 minutes between trains. That way a Metro train would signify a frequent all week service, similar to how trams do now. Prospects and priorities post-Metro Tunnel (and the next state election) are discussed  next. 

2026 and beyond

The minister at Estimates encouragingly last week said that the 2025 budget service improvements was just the 'first stage' of Metro service uplifts. This may indicate that there's more to come in the (pre-election) 2026 state budget. This could mark a welcome revival of political interest in public transport service, noting that neither the 2018 nor 2022 election campaigns had significant metropolitan rail frequency promises from either major party. 

Ms Williams also said that infrastructure was not something built for its own sake but as a means to ends like service uplifts. This is sensible; having infrastructure construction subservient to other goals is essential to ensure the right things get built and maximum value is extracted from what we do build. As opposed to setting your mind on a particular project before clarifying the problem it is meant to solve. 

Before we think about next year, let's go back a decade or so. 

If not quite 'projects first', Labor's 2014 policy was certainly 'project jobs first'. This had its genesis in Project 10000, a 2013 plan to create 10000 construction jobs hatched when in opposition. As well as pleasing key unions this reassured the construction industry that there would be continued work, especially desired as Labor needed an alternative to the Liberals' East-West Link.

The emphasis was always on the jobs and building; the benefits were a welcome politically saleable by-product. Project 10000 was realised as level crossing removals, the Metro Tunnel, West Gate Distributor and other road upgrades. More of the latter were added, including the West Gate Tunnel, North East Link and further level crossing removals when in office. The enthusiasm for major projects peaked in 2018 when the program was packaged as the Big Build, culminating in the announcement of the Suburban Rail Loop a few months before that year's state election.

While projects were popular there were not always evaluations of whether similar benefits were possible for less money with a different suite of projects. Also many opportunities to improve service or increase active transport connectivity that were not dependent on big infrastructure builds were either not taken or got deferred, possibly to magnify projects' BCRs through artful bundling. 

Furthermore, low interest rates made major capital projects both financially and politically attractive, especially relative to ongoing budget spending on services. Hence major works was not just a Victorian Labor thing; the same conditions encouraged similar mega-projects in other states too. Although Sydney and Perth were better at doing both service and infrastructure than Brisbane and Melbourne have been, thus maximising benefits.  

Today's situation is different. Higher interest rates, tighter government finances, soaring construction costs and increased interest in non-transport construction, notably housing, has made transport megaprojects harder to justify. The transport construction frenzy has given us a huge amount of relatively new but still underutilised assets, notably on the rail network. So now is a good time to work them hard to realise their benefits. 

With a start made on 'more trains more often' in the 2025 state budget and the Metro Tunnel operating, what should 2026's budget feature to spread service uplifts to more lines?

Here's some tips, starting with the cheaper and easier uplifts that cut the longest waits first. 

Smaller

* 20 min maximum waits on the Sandringham line: Boost Sunday mornings from 40 to 20 min and an early Saturday morning boost (if not already done with the Metro Tunnel timetable). Approx 10 more services per week required.  

* 20 min maximum waits to Mernda and Eltham: Boosts Sunday morning service on each line from every 40 to every 20 min, evenings from 30 to 20 min. Similar package as 2025 funded for Craigieburn and Upfield involving 60-70 extra weekly services per line. This uplift would give all stations to Clifton Hill a 10 minute or better service from 7am to midnight 7 days. 

* Shoulder peak weekday upgrades on the Craigieburn, Werribee and Mernda lines: Increase the hours per day a frequent service applies, starting with busiest lines.  

Larger

* Greenfields timetable for Burnley group: A major revamp including (a) fewer peak stopping patterns (b) 20 min maximum interpeak weekday waits for Belgrave and Lilydale, (c) Sunday morning boosts from 30 to 20 min, (d) Reduced evening maximum waits from 30 to 20 min on all lines, (e) 10 min service to Glen Waverley (daytime) and Ringwood (day and night). 

* Weekday interpeak and weekend frequency upgrades from 20 to 10 min on Craigieburn, Werribee, Mernda, inner Hurstbridge and Upfield lines, roughly in that order, building on previous shoulder peak upgrades. 

* Finishing the job on Metro Tunnel upgrades if the Enhanced service option is not already running.

* V/Line service uplifts including upgrading Melton to every 20 min weekends, higher Wyndham Vale frequencies and improved Seymour line service. 

* A complementary rail infrastructure program driven by service and capacity needs for the west, north and outer south-east. (the need to boost service, then enabling infrastructure, then service again was cited by the minister in Estimates)

Conclusion

To summarise, Metro (as opposed to V/Line) train service has lagged population growth in the current government's first decade. This has led to Melbourne having rail service levels inferior to that of other cities, with frequent service largely confined to peaks on most lines.

Long waits at night and weekend mornings also reduce the ability of trains to support major events and commutes for retail, food, hospitality, health and event workers.

There is also a large geographic inequality, unrelated to patronage, between busy poorly serviced lines like Craigieburn and highly serviced lines like Frankston. 

The 2025 state budget has funded welcome service upgrades we know about on some lines with others (we don't yet know) coming on the Metro Tunnel lines.

Also welcome is the minister flagging the possibility of further Metro service upgrades in 2026. It is hoped that these will be spread widely across more of Melbourne, including the Clifton Hill and Burnley groups which were out of scope of the Metro Tunnel related service upgrades. 

A robust program of rail service uplifts will enable the government to tell a good story on both rail infrastructure and service, especially if investment in the latter enables a substantial per capita rebound.   


Index to other Timetable Tuesday items here