UN 236: Auditor General reports on bus planning

Today is exactly 5 years (or 1831 days) since Victoria's Bus Plan was released. Possibly not uncoincidentally the Victorian Auditor-General (VAGO) had a report critical of bus network planning in Victoria tabled in parliament yesterday. It makes interesting reading. 

VAGO found that the Department of Transport and Planning had an ambitious vision for buses to be a mass transit option by 2031 but that it could not be achieved given the current state of the network and slow progress. 

Audit also found some performance measures differed from customer experiences, a failure to meet many targets and poor quality data including annual bus patronage being undercounted by about 25 per cent (with wide variations). Some, such as the practicality of measuring punctuality at every stop, were contested by the department (which on this matter I'm inclined to agree with). 

Furthermore, DTP was criticised for not being open with Victorians on Bus Plan progress. Reporting was possibly made harder or more embarrassing by the plan's flagship network reforms being predicated on funding that the Department was unable to win from the budget process or find from internal sources.  

The Department partly accepted or accepted in principle the audit's three recommendations. And it has promised to release a map of main bus corridors through the Plan for Victoria process and use the gap between actual and desired services along them to inform its investment pipeline (that gets put to government for budget funding). 

The audit took 11 months and cost $720 000.  

Before we get onto the audit itself (about which I've made a video), here's some context. 

Background - main events in bus planning

In a nutshell, the state government released a not detailed bus plan in 2021, hired a bus reform team, commissioned consultants reports, developed a detailed Bus Network Reform Implementation Approach, proposed main road buses every 10 minutes and promised bus reform in Melbourne's north and east with public engagement before the 2022 state election. The pilot parts of the scheme were to start in 2025 with a staged plan to reform buses across the rest of Melbourne by 2031. It would have been transformative if it happened, with patronage up tens of millions (making it bigger than projects like North-East Link or Metro Tunnel in moving people). 

 

Safely re-elected, the government effectively scrapped the plan with the 2023 and 2024 state budgets bringing no joy for bus reform in the target areas. However the 2025 and 2026 state budgets saw a revival of interest in buses with piecemeal bus upgrades funded, including some in Melbourne's outer north. While the scale was smaller and the approach was different, the government still sold these uplifts as being part of the Bus Plan.  

This reframing has disappointed the Sustainable Cities Better Buses campaign which supports wholesale rather than piecemeal bus network reform in Melbourne's west. This would have delivered buses every 10 minutes along main roads similar to the grid network envisaged in the Bus Network Reform Implementation Approach. 

This non-performance and a failure to keep Victorians up to date did not get past the Auditor-General in their largely critical report.  

Bus Plan Audit - My video analysis

VAGO's data dashboard on bus service quality

In addition to the pdf report, VAGO has a data dashboard on its website. If you view it full screen you will find extra information such as interactive maps that let you search access to what they define as good quality transport by postcode and suburb.

In my view VAGO overestimates the prevalence of high quality transport by avoiding having service standards on weekends (apart from the route running 7 days). As a minimum they should have used the still influential 2006 MOTC 9pm finish on weekends and not just weekdays. And preferably also had a frequency standard on weekends similar to weekdays. If these were applied the number of stops considered to have high quality bus service would fall greatly. 

Another distortion that causes VAGO to overestimate the amount of high quality service (even on its loose criteria) comes from frequency averaging. Where a route is relatively frequent in the peaks but infrequent off-peak it may still have an average frequency that tips it over the 20 minute threshold despite its long midday gaps. A key example is the popular 495 in Point Cook that VAGO counts as high quality despite its 40 minute midday weekday headway. 364 in Warrandyte, 385 and 388 around Doreen, 432 in Altona North, 463 in Hillside and 476 in Keilor have similar issues.    

However VAGO also underestimate service on stops served by multiple routes because they take the easy way out of considering data by route rather than on a network basis. A more rigorous method would likely show that corridors like Belmore Rd (Routes 302/304) and much of Altona Meadows (411/412) meet their criteria for high service. VAGO's single route counting differs from customer experience which in cases like the above overwhelmingly treat the routes as a combined more frequent service (even if Transport Victoria isn't always great at describing this). 

If the auditors are going to criticise DTP on their data (which they did for patronage), then their own methods need to be above reproach too. 

What happened when

It should be noted that auditors-general are restricted in the scope of their inquiries. To quote Yes Minister they are interested in the "administration of policy" rather than the "policy of administration". Auditors also do not comment on the appropriateness of policy or political priorities, unlike what I can do here.

Neither VAGO nor myself have access to cabinet in confidence documents, though I have freedom to draw longer bows. This is important because the Bus Plan was substantially unfunded when it came out.

Anything arising from it would be subject to DTP's ability to win budget funding. Despite benefits in terms of number of trips improved or made possible being bigger than 'Big Build' projects like North-East Link or the Metro Tunnel, bus reform never had the glamour of those projects.

This may be attributable to weaknesses on the part of previous DTP bosses in being unable to make the case for funding, a government focused on major capital projects to the exclusion of service uplifts and/or political concern over the elevated risks of the trade-offs inherent in any cost-effective large-scale bus network reform. As transport minister the current premier sought to avoid controversy on network reform matters as you can see here and here.  

Important bus reform milestones in the last five years are as follows: 

June 2021: Minister Ben Carroll releases Victoria's Bus Plan. It has some good principles. But it is unfunded and has little substance, relying on a future Bus Reform Implementation Plan for detail. 

September 2022: State Government announces major bus reviews with Melbourne's North, Melbourne's North-East and Mildura as pilot areas. First round of public consultation is held. 

November 2022: State government returned at election with strong majority. 

May 2023: State Budget funds little new for buses. No sign of Bus Reform Implementation Plan or progress of bus reviews. 

September 2023: Change of premier and transport minister (to Jacinta Allan/Gabrielle Williams)

Late 2023: Cabinet meeting discusses bus reform. Appears to abandon it (not known at the time due to cabinet confidentiality).

March 2024: Trung Luu MP moves Legislative Council motion for internal documents on bus planning to be released. 

May 2024: State Budget funds little new for buses (Route 800 upgrade only with any other budget submissions from DTP rejected). 

May 2025: State Budget funds some upgrades for bus services in Melbourne's outer west and outer north but not the systemic network reform envisaged in the Bus Reform Implementation Plan. Government reframes the language of the Bus Plan, saying anything it does with buses is consistent with the Bus Plan despite the abandonment of the Bus Reform Implementation Plan.  

September 2025: Bus reform documents tabled in Legislative Council. These demonstrate the progress and then abandonment of bus network reform in the large-scale style envisaged. 

May 2026: State Budget funds further bus service uplifts ($100m over 4 years). These concentrate on improved weekend services including on some main routes. These include some routes identified in the Bus Reform Implementation Approach but frequency uplifts are typically to every 20 rather than every 10 min. 

June 2026: Minister Gabrielle Williams uses social media to ask Victorians for their ideas on improved bus services. 

June 2026: Auditor-General's report on Victoria's Bus Plan tabled in Parliament.  

Audit approach and outcomes

VAGO's report on Bus Plan execution by DTP has not surprisingly been critical of the department.

However there are aspects of the audit that I'd have preferred were different. For instance VAGO's lax definition of high quality service reflects more on the geographic and class biases of white collar public servants than actual bus user needs (who need more weekend service as the government itself is starting to recognise). Their sample of routes to audit for coordination wasn't ideal. It was silly to use 2021 as a baseline for punctuality, notwithstanding the footnote. I'm inclined to agree with DTP on punctuality performance measures but if VAGO were serious they'd be probing progress on bus priority and run time revisions to address the causes of lateness. More on this in the video above.   

The map of strategic bus corridors that DTP says it will publish under Plan Melbourne will be  welcome. And the department using it to assess gaps between aspirational and current service to produce a pipeline of service improvements along major corridors is super-important. 

Some of these will require funding for new service kilometres while others need a willingness to tackle inefficient overlaps to form more frequent routes from existing resources. A strategic corridor bus network map should help DTP focus on this harder than it might have in the past (with the result of us being some local buses still running at 11pm but while busier routes on key corridors are still less frequent at night and/or with early finishes). 

As learned from the ill-fated North, North-Eastern and Mildura bus reviews, the critical point to success is political will, preferably on a larger scale than the good but still limited uplifts this year's state budget enabled. 

See other Useful Network items

TT 232: South-east Metro Tunnel bus recoordination and Mornington Peninsula network changes

Bus Recoordination in the south-east

The third (and I believe) final tranche of bus timetable recoordination to reflect the February 1 2026 Metro Tunnel and related train timetables will come into effect on Sunday July 5. 

The previous stage, recoordinating bus routes in Melbourne's north and west, happened in April, with details in the Projects section of the Transport Victoria website here. 

Recoordinations are typically minor tweaks of a few minutes to optimise connections (and sometimes bus running times) but a few lucky routes (like 467 and 503 in April) can get more trips to match bus with train frequencies. This is important because Melbourne's north (especially) still has many bus routes that run every 22-30 minutes in areas where trains are normally every 20 minutes, with resultant haphazard connections.  

As foreshadowed in the item on the April 2026 bus recoordination, more time changes were to come, but this time for buses feeding the Pakenham, Cranbourne and Frankston lines. These start on Sunday July 5, with a list of the 31 routes involved here. (627, 701,705, 708, 709, 760, 767, 770, 771, 772, 773, 774, 775, 776, 778, 779, 780, 782, 783, 788, 811, 812, 824, 825, 828, 832, 833, 857, 858, 863 & 887 for those interested) 

As far as I can tell the changes are small except for Route 833 between Carrum and Frankston. This route (written as 883 in the heading but corrected yesterday) will have its Saturday frequency boosted from every 60 to every 40 minutes.

This gets particularly strong use between Frankston and Carrum Downs so the upgrade will be welcomed. It does however mean that there will be uneven intervals between buses at locations such as along McLeod Rd, Patterson Lakes as the 708 that overlaps it will remain hourly on weekends. The same applies in areas where the (also hourly on weekends) 832 is an option for travel to Frankston.

As you can see from the print just above, this recoordination item is in the News (not the Projects) section of the TV website (unlike the April recoordination announcement). I prefer this but consistency that breeds repeated behaviour and thus reader familiarity and trust would have been nice.    

Also the south-east item lists Mornington Peninsula routes 781, 784 and 785. It needn't have as there is no elaboration of changes affecting them with the drop-down route list not including them. Instead that's the job of a separate website item, discussed below.  

Mornington Peninsula bus network reform

Also starting on July 5 are reformed routes on the Mornington Peninsula. Most notable features include: 

* A new cross-peninsula Route 886 between Mornington and Hastings, operating hourly all week

This will enable travel between both sides of the peninsula without the need to backtrack to Frankston. While Transport Victoria don't really promote buses for day trips or tourism, the 886 will make it easy to visit attractions across the peninsula such as Mornington and Tyabb's antique markets in a single day trip. 

* Deletion of Route 786 between Rye and St Andrews Beach due to low patronage

Removes all public transport network coverage from St Andrews on the Bass Strait side of the peninsula. 786 is the quietest regular bus route on the network with under 2 passenger boardings per service hour recorded in 2022. This is similar to that recorded on the old Route 687 to Chum Creek  before it got deleted as part of Healesville's bus reform a few years ago. Its service hours are being reused on other Mornington Peninsula bus routes.    

* Rerouting of bus routes 784 and 785 in the Mornington area

Reduces overlaps between routes and speed travel to Frankston from Mornington East. However this is at the cost of making buses more complex in central Mornington, something TV should be explaining better (more on this later).  

* Longer operating hours on bus routes 781, 784 and 785 

Extra trips so buses run from approximately 6am to 10 or 11pm across the week. Wider operating hours including later night trips has been a welcome and widespread trend of this and other recent bus network upgrades. However their application has been uneven; Melbourne's highest-usage routes such as the 900, 901, 902 and 903 SmartBuses now finish earlier on Sundays (around 9pm) than many local routes.  
Passenger communication

The Transport Victoria news item doesn't explain these network changes very well. There are links to  pdf timetables that show individual route maps but nothing for the network as a whole. That's important as some peoples' nearest bus route will change or they may need to wait at a different stop to reach their destination.

PTV and Transport Victoria communication has always been heavily text based. This limits the inclusiveness of their messaging, especially for people whose first language is not English or are visual thinkers. This bias cannot be put down to a lack of internal resources; DTP already makes maps for consultation purposes but forgets they exist a year or two later when they could help explain a reformed network.   

DTP has so many layers of management that they are forced to spend disproportionate time in meetings (it's a mathematical fact - work it out on paper - meetings rise exponentially with executive numbers). If results are any guide, this distracts bosses from actual productive work such as looking at outcomes (such as community benefits), processes (like knowledge management that  maximises value from work already done) and outputs (such as its website) to drive improvements. 

Central Mornington stops

What about central Mornington? You could argue that it is generously served with 5 buses per hour to Frankston on weekdays and not much less than that on weekends. However it is a key destination and people need to know where to catch their bus from given that (a) service is spread over four mostly hourly routes and (b) the altered routes change things. 

The complexity of buses in central Mornington was considered a problem about 20 years ago. This led to central Mornington bus stops being simplified about 12 to 15 years ago following state government funding in 2011 (background here, here and here).

That meant that you could get all local routes (781, 784, 785) towards Frankston from the one stop, even if the buses had to deviate a little. All three routes were relatively direct to Frankston so you would typically board the first bus that came along. Ideally scheduling would permit as close as possible to a 20 minute combined maximum wait given that each of these routes runs hourly. 

Quicker travel via the 788 from the Bays Hospital stop was possible but until improvements a few years ago (every 40 to every 30 min on weekdays, 80 to 40 min weekends) its frequency meant that unless you timed it well the wait time often exceeded (or felt like it exceeded) what you'd save by taking the faster 788.

The map below shows the main stops in central Mornington as they currently stand.  

As noted before Transport Victoria prefers text over maps. They resist drawing new maps and won't necessarily even reuse ones they previously made.

To fill this gap I made one for what I think will happen in Central Mornington, the busiest stops on the network outside of Frankston. Undoing the 2011 consolidation, travel from there to Frankston will be less legible with three rather than two stops. Rather than just rock up at the stop on the north side of Barkly St as you probably would today, this setup gives a bit of a nudge to check times and possibly walk a block to catch the faster route 788. 

This is the new network's main trade-off, noting that 784's new alignment delivers substantial travel time savings for Mornington East (which previously had no fast option to Frankston). Also people need to be made aware that the new 886 to Hastings will depart from the Bays Hospital, not the previously main stops in Barkly St. 

This needs a bit of explaining so that people get the right stop. Noting that these are well used stops and the area has a senior population skew, with a higher than average proportion of less mobile passengers. Hopefully DTP/Transport Victoria will come to appreciate this with graphical descriptions of the new central Mornington bus stop arrangements provided both online and at the site.  

See other Timetable Tuesday items here

UN 235: PTUA's plan for Geelong and Bellarine buses

Today I'm showcasing another vision for improved bus services, this time for Geelong, Bellarine Peninsula and Bannockburn from the Public Transport Users Association Geelong Branch. 

It is in response to the bus review DTP is conducting for the area, following funding in a recent state budget. They asked for public input back in April and PTUA responded with a detailed submission. 

My personal view is that Geelong is one of the places that least needs a wholesale bus network review. Geelong itself got a major network review in 2015 with routes dramatically straightened and (mostly) upgraded to every 20 minutes off-peak on weekdays. That compares well with routes in metropolitan Melbourne and meshes with trains that run every 20 minutes. 

If you want to look at places most in need of an overhauled bus network look at Wodonga, Shepparton and Mildura. Infrastructure Victoria recommended revamped bus networks in these places and more. The government actually started a review for Mildura (when it had an independent MP) but failed to carry through after it was safely returned in the 2022 state election. 

Having said that the Geelong area still needs some bus service improvements. Despite generally better weekday frequencies than Melbourne suburbs operating spans are shorter, especially on weekends. Late starts, early finishes and low frequencies stymie travel to places like Queenscliff, especially on weekends when intending visitor numbers are highest. And growing Bannockburn needs much more than its current sparse service. 

The PTUA submission tackles all these by recommending higher frequencies and longer hours. It also supports bus priority, improved bus shelters and the retention of the Moorabool Street bus interchange in the Geelong CBD. The latter is a hot topic locally; some retailers want the bus interchange moved due to antisocial behaviour and the sorts of people they say it attracts. A trial of a Night Network type service on the main bus routes is also advocated. 

Geelong's train service used to be hourly. It has been every 20 minutes seven days per week since 2024 following successive service upgrades. Line usage has surged, largely due to urban growth in Tarneit and Wyndham Vale with more to come when the new Tarneit West station opens in a few months.

However its peak timetable remains with excessively complex stopping patterns. Also a short-sighted decision was made to retain 40 minute weekday interpeak frequencies for Waurn Ponds despite a 20 minute service now operating on weekends. These oddities undermine the potential role of rail to operate as a spine for fast local travel for some within-Geelong trips. 

The key decision taken when planning Geelong's bus network is its base frequency. As noted before this is commonly 20 minutes on weekdays for local bus routes. It's on weekends when gaps widen. Even Geelong's busiest route (Route 1) has a (slightly uneven) 30 minute headway on weekends. This is actually similar to some Melbourne SmartBuses but was instituted when Geelong weekend trains were every 60 minutes. Buses were not significantly upgraded when weekend trains went to 40 and then 20 minutes. 

If you are not going to have buses running at turn-up-and-go frequencies and you value connectivity with trains then the other option is a timed transfer network where there is a family of frequencies that evenly mesh with train services. For example if trains are every 15 minutes then buses might be every 15 minutes for main routes, every 30 minutes for middle importance routes and every 60 minutes for local or semi-rural routes. 

In Geelong's case the pulse is set by a 20 minute train headway to South Geelong. The menu of acceptable frequencies for buses then becomes 20, 40 and 60 minutes (though 40 won't mesh with 60 for even bus to bus connections). A 20 minute interval with most routes (ie matching train frequencies) is basically what the planners in 2015 went with for weekday services. In contrast the 2014 Brimbank and 2015 Wyndham networks planned at a similar time went with a harmonised hierarchy with main routes every 20 minutes and local routes every 40 minutes. Melbourne uses 40 minute frequencies more than anywhere else in Australia. Its advantage is that it's better than 60 minutes but is not a memorable clockface headway. 

The PTUA submission addresses the headway harmonisation problem by ignoring it. Their proposed bus network is based on service every 15 minutes for urban Geelong and Lara routes and every 30 minutes for peripheral routes. Opting for this produces a memory timetable good for local trips but sacrifices even connectivity with trains, especially in cases where buses every 30 minutes meet trains every 40 minutes (or less worse every 20 minutes). It is possibly true that outside main commuting times travel within a city is more significant than trips involving a train connection. However longer tourist type trips where people are travelling to locations like the Bellarine Peninsula may have a significant rail connection component. But PTUA's headway choice does avoid horrid 40 minute bus intervals for outlying areas which might have been its main reasoning. 

The submission correctly (in my view) says the structure of network is strong. Thus it does not suggest major overhauls for the core of the network. But it does for some outlying areas with maps provided. 

Anyway that's my summary. Have a read and let me know in the comments what you think. 

See other Useful Network items