Sunday, December 01, 2019

Scotched at last minute: How Transdev's 2015 Greenfields bus network didn't happen

Today marks five years after the public was told about what was to be the most audacious 'big bang' changes to Melbourne's major bus routes in years. 


In 2013 French-owned bus operator Transdev won the right to run a third of Melbourne's bus network from local operator groupings including Melbourne Bus Link and Ventura Bus Lines. Other smaller local bus companies, as represented by BusVic, were also not happy. They feared routes they operated might be next. And it was widely thought that Transdev put in a low-margin bid to win the business. 

The Coalition government, wanting to save money, initiated the spill. It wasn't publicly controversial like previous transport franchising as the routes were already run by private rather than government operators. And it concerned buses, not trains or trams. Officials intimated that incumbent bus operators were on fat margins and competitive franchising could reduce costs. And the Auditor-general confirmed a $33m saving with Transdev's contract.   

Good things

Transdev Melbourne brought in some improvements. For example the new franchise contract standardised public holiday arrangements for the routes they ran. Before that it was confusing knowing which route ran which timetable on a public holiday. 

Then there is bus run times. Routes inherited from Melbourne Bus Link often had timetables unchanged for years (browse them on Krustylink). Run times were unrealistic and late running was chronic. While still chided for lateness, Transdev Melbourne has steadily updated timetables to reflect modern traffic conditions. This would not have been cheap; more run time means more bus hours, more drivers, and more buses just to maintain an existing headway. Good pre-contract due diligence would have picked up the need for this and informed bid costings. 

Transdev Melbourne also deserves credit for its active approach to network reform. The network it had taken over (again especially the Melbourne Bus Link portion) was frozen in time. Parts were overserviced while other parts were underserviced. Like run times service levels did not reflect modern travel demands. 

Major network changes came to some areas in mid-2014. Service levels were insufficient on some routes but the general thrust of most changes was sound. Overall the network was simpler afterwards than before. Fishermans Bend was a major success with higher patronage requiring a subsequent frequency upgrade. Read more here. The August 2014 Table Talk has a detailed summary of the 2014 changes

Bad things 

Unfortunately cheapness brought nastiness rather than value. Standards slipped. Cancellations surged. Transdev Melbourne buses invariably came with layer-upon-layer of grime and graffiti. Being too tight to hire enough professional cleaners, Transdev leaned on staff to clean buses in unpaid time.

It wasn't just cleaning that was being palmed off. Safety too. 2017 was a horror year with a frontless bus carrying passengers and much of the fleet (more than the number reported) being deemed unroadworthy by Transport Safety Victoria. The (now Labor) state government responded by cutting short Transdev Melbourne's contract extension and continuing with internal restructuring that would fold PTV (who managed the contract) into the Department of Transport with greater ministerial oversight. 

Back in 2014 though, it seemed that only riders were seeing the problems. Bus crises in Melbourne don't make the front pages as much as rail meltdowns. Public Transport Victoria proved more a lap dog than a watch dog when it came to managing the contract as later events and an auditor-general report showed.

In this industry key managers frequently swap from poacher to gamekeeper and back so points might not have been pressed harder than was gentlemanly. Especially if there was the risk of a National Express-style walkout and budget blowout if replacement operators demand more to do the job properly. Someone would then have to own up that, like with the first round of rail franchising twenty years ago, they banked on savings that proved illusory.

The 2015 'Greenfields network' proposal

While radical by previous standards, the 2014 network change was just a warm up for what Transdev was planning for the following year. This was to be the promised greenfields network and timetable due for April 2015. It would affect most routes Transdev ran, including the orbitals and those untouched by the 2014 reforms.

Passengers first saw an official announcement on December 1, 2014. Visit the Web Archive link on the Wayback Machine to see how Transdev presented it to the public. Enthusiast discussion about it appears on Bus Australia here

Not unrelated (for reasons that will later become clear) was the 2014 state election. This took place two days prior on November 29. The Coalition government (that hired Transdev under its bus franchising policy) lost to Labor which campaigned on a program of building infrastructure, reviving technical education and an extra public holiday.  

The close timing of when the network proposals became public and the election would not have been a coincidence. Especially if it appeared to be at or close to a 'cost neutral' network change with upgrades paid for by cuts elsewhere. And, as would later become apparent, it would have many 'nasties', especially in the northern and western suburbs. 

The initial concept

Even before details of the new Transdev network were out, it was known there was going to be one. After all it was foreshadowed back when Transdev Melbourne started in 2013

A conceptual network was shared with affected local governments as early as April 2014. At least some information made it into published meeting records (eg Glen Eira 8 April 2014) that could be viewed online at the time (though Glen Eira removes very old meeting minutes from its site so it's no longer available). However you can read a summary of it in May 2014 Table Talk. There's also some enthusiast discussion here. That discussion ended with a response from Transdev confirming the very draft status of the network proposal (and also printed in August 2014 Table Talk). 

The public version

Today's anniversary is based on when the revised version became public - ie 1 December 2014, two days after the state election. Look at the Wayback Machine links and read/download the pdf brochures by region (via below). 

Summary of changes

It's hard to do justice to the reformed network here but the general thrust of the changes included: 

* Reform the SmartBus orbitalsSplitting them into a series of shorter routes, some with amended origins and destinations. A map is below. 

* Change service levels of the split sections of the orbitals. Instead of a flat 15 minute weekday/30 minute weekend service timetables were altered so that some routes became more frequent (as high as 10 minutes during peaks) and some less so (down to every 40 minutes). Overall there was a  service reduction in the north and west and an increase in the south and east. Claims were made that there would be better connections with trains but this was not always so. 

* Western suburbs: Simplification of 216/220 between Sunshine and Footscray (eventually done just two weeks ago in slightly different form). Some routes would be upgraded (eg 220 Sunshine - Footscray) while others (eg 216, part of 220 and 223) would have reduced off-peak service. Also cuts to SmartBus service due to split routes. 

* Northern suburbs: Splitting of SmartBus orbital routes and frequency cuts on most routes. The split at Northland would have been be particularly unhelpful for through passengers with incompatible frequencies guaranteeing poor connections on weekends. No attempt was made to consolidate service with parallel routes run by other operators (eg 513 and 527 between Coburg and Northland). However some destinations were changed, for example the direct SmartBus from Greensborough ran to Melbourne Airport. 

* Eastern suburbs: Generally increased services on orbitals, particularly during peak times and on weekends but with some route changes including a truncation of Route 901. Major changes to local routes in Manningham area with a generally simplified network. Route 907 gains a large frequency boost while others are modified or downgraded. 

* Southern suburbs: Generally increased services on orbitals, particularly during peak times and on weekends. Simplification of routes in Brighton area (partially done two weeks ago in different form).  

* City area: Involves changes due to network reforms in east. Also split of 216/219/220 in city and other reforms (done in a different way since and made permanent from 2 weeks ago). 

Overall the changes were major. Some trips became better while others became worse. There was a transfer of resources from the west to the east with little credit given to boosting busy routes in the west. And the network did not necessarily touch quiet routes in the south and east, for example the 922/923 around Sandringham.

In other cases it even added buses to quiet sections. For example the proposed network delivered a 10 minute Saturday service to the 903 all the way to Mordialloc despite the Mentone - Mordialloc segment being quiet. The same could be said for sparsely populated areas in the north-east between South Morang and Greensborough still having a SmartBus and two still running from there to Templestowe. 

Public engagement

There were two mechanisms for this: 

1. An online Survey Monkey survey

2. What was described as Community Information Sessions

Note 'Information' not 'Consultation'. The thrust was very much one-way. That is them telling you about the new network and for them to answer questions on it. While you could give feedback there was no mention that it would be used. 

There was more contact with the public than the 2014 changes, which were introduced without consultation. However 2015's consultation style was criticised by some as a 'going through the motions' exercise. These concerns did not go unnoticed by the new Labor government as you'll see later. 


Unusually for buses, the new network got significant media attention.  While the facts were more nuanced (with cuts and increases on both sides of the Yarra) the summary that it was a 'take from the west, give to the east' change was broadly correct. 

This message about 'giving to the east' was confirmed in the comments from Transdev contained in that same article.

What was missing were the reasons for the poor patronage performance claimed for northern and western parts of the SmartBus routes. Some might include: 

* Poor/sparse/low-density catchment. Affects both the Sunshine - Altona North part of the 903 and the South Morang - Greensborough portion of the 901. Yes, these might not deserve a SmartBus service level. The risk here is they might depress averages for adjacent parts of the route including those that perform well with some wrong conclusions being drawn.  

* Overlaps between SmartBus and regular routes. The busier parts of SmartBus routes in the south-east are generally free of large overlaps. But this is not the case elsewhere. Example overlaps include 902 with 566 west of Greensborough, 903 with 513 and 527 between Heidelberg and Coburg, 903 with 465 west of Essendon and 903 with 232 and 411 in Altona North. Fixing these overlaps would have required a multi-operator approach that Transdev by itself would be unable to implement. An example of this type of network reform and what it can achieve is the Murray Rd Megabus described here.  

* Overlaps between SmartBuses themselves. This is most notable in the north-east, with an overlap between 901 and 902 in the Templestowe area. Transdev's proposed network appeared not to address this. 

* Veering away from major trip generators. In the east it would be unthinkable for an orbital route to miss a major shopping centre that is reasonably in its line. However this is what happens with the 903 that skips Highpoint in favour of serving the far smaller (and more local) Milleara Shopping Centre while duplicating a local route (465). Both the weak trip generator and the duplication rob the orbital of patronage in the west.  

If you removed those reasons you'd get improved productivity from the orbitals in the north and west. The problem then is that there would have been no rationale to cut their service levels. This would have made it harder to free resources to boost services in the east (which happens to have more marginal seats). 

Apart from the various overlaps with the 903 orbital in Melbourne's west and north, most instances of service inefficiency on Melbourne's bus network are in an arc from the north-east to the south-east. However releasing resources from them to put into eastern suburbs service upgrades would have required a wider network planning approach involving more than just Transdev. 


The November 2014 election result turned out to be fateful for Transdev's Greenfield network. 

April 2015 rolled around. 

The following government release appeared. 

Transdev's network proposal was dead. Transport Minister Jacinta Allan vetoed it. 

Too many people would have lost out had it gone ahead. 

And a shot was taken at the limited consultation.

Transdev would have been shocked. Assumptions that the government would largely leave them to do their own thing (as Minister Mulder might have done) were turned on their head. While the Transdev Melbourne Franchise Agreement hadn't changed, its customer's philosophy had. Instead of being a commercial franchisee with substantial freedom to make business decisions such as network reform (with the risks and rewards this entails), Transdev had effectively become a government contractor (more like the other bus operators) running an inherited network it couldn't easily change.

While the new network had good points I think on balance the minister was right to veto. The new timetable would have shafted the west and north with few compensatory upgrades for busier routes. This is because there wasn't the sort of network-wide view that could have cut inefficiencies and delivered service gains across Melbourne like we often discuss here.

What about the "more balanced bus network proposal" promised? 

Nothing has been heard of that in the four or so years since, although that didn't stop people asking

There appeared no progress when a similar vague answer was given in parliament seven months later.

Indeed it's been slim pickings for followers of bus reform in Melbourne. 

Our city's grown and changed but few bus routes and timetables have

What could have happened

PTV was the Liberals creation, although The Greens also supported the concept. I'm not sure if Labor ever warmed to the concept of PTV as an independent agency (the 'it wasn't our idea' syndrome). Especially if it was an agency that outsourced some of its planning to a franchise operators with less ministerial oversight compared to if it was done in house.

Such oversight appeared to have been desired more because Labor wanted to prevent politically controversial network reforms happening rather than because it had its own vision for the network. Apart from delivering routes promised in its 2014 policy platform Labor had none; its other priorities in public transport mostly being infrastructure rather than service-based.

Under a more service-oriented government PTV might have got a greater role in planning (as it did for networks like Wyndham and Brimbank). Then it could have reformed networks by area, involving both Transdev and non-Transdev routes. And time-staggering implementation, such as was done during the minimum standards bus upgrades, would have meant a smoother workflow in areas such as data processing, signage, stop timetables and communication.

Such multi-operator area-based network reviews would have produced better results with an elimination of inefficient duplication between Transdev and non-Transdev routes. Resources could have been retained in the north and west while over-servicing in the south and east could have been reined in, with resources diverted to busier routes in the area where pressing needs existed.

The future

It hasn't been a good last three or so years for advocates and followers of bus reform in Melbourne. With other transport projects on the go, the Labor government hasn't been interested. And any internal advocacy in favour of bus service and network reform from PTV (and later the Department of Transport) has either been weak or ineffectual.

Non-government bus service campaigns were not strong with no 'what's in it for me' type of detail to get people excited. Also, not unrelated, is that promises made before the 2018 election with regards to buses were either weak (Labor) or last-minute and lacking in detail (Coalition).

However there are signs of stirring and even modest action. 

A new minister with a bigger government majority might be more game to do stuff. On the other hand finances are tighter, although this just make it more critical that over-serviced parts of the network are pruned

Some extra peak services now run on some bus routes in Fishermans Bend and Doncaster. Route 246 got a Sunday upgrade. And small network reforms started just two weeks ago, affecting Transdev routes in Sunshine and Brighton. 

They're a far cry from the bold network transformation we had in the 2013 to 2016 period. 

But they're better than nothing and there's possibly more hope for bus reform now than twelve months ago with talk around the traps that there may now be an ambitious bus patronage target.

Melbourne on Transit bookshop

Favourably reviewed books about transport and cities. Purchases via these links support this blog and its independent reporting (at no extra cost to you).   


Craig Halsall said...

Overall a good summary of the failed Greenfields plan - however you might want to re-read their plans for the south-east network more closely as it has more nasties beyond network refinement in Brighton.

Transdev planned to scrap the link linking St Kilda and Sandringham (they felt changing to the 246 at Elsternwck is sufficient), as well as abolish the 922 & 923 in Beaumaris altogether, other a couple of school time deviations along Reserve Rd & Weatherall Rd. The result also see notable cuts to the frequency of buses through the area, as the resources were not pumped back into the simplified 600 offering. Late night services were also to be deleted.

A former ALP transport minister is known to live in Black Rock, I wonder if he had any say in knocking back the changes during cabinet room discussions?

Peter Parker said...

This is true. Brighton was an example where they couldn't do things in conjunction with other operators - eg fewer but simpler east-west connections or reforming missing links eg Nepean Hwy/823 or Hampton St in exchange for downgrading some of the low usage coastal routes. It was a highly unsaleable network for the area as I don't think there were any good things to offset the bad. Had they taken a bit less out they could have had a simpler network. Unfortunately the 15 min (as opposed to 10 min) frequency of the Sandringham line reduces the range of coordinating frequencies. Eg what is now the 600/922/923 could have been dropped to 2 routes every 40 min each for a combined 20 min frequency, possibly run 7 days/week.