End of year blog clearout

It’s the end of the year, so I’m going to clear out some blog drafts, snippets of writing, going back a bit over a year.

The Great Australian Transport Dream

If the Great Australian Dream is home ownership, preferably a quarter-acre block with a huge backyard, then the Great Australian Transport Dream might be free-flowing traffic everywhere you go, and a car spot right out the front of wherever you’re visiting.

Both dreams are dying, because they depend on space which is no longer available as our cities continue to grow.

In fact arguably if you want either or both of these, you need to move to the country, to a town or smaller city.

And yet our politicians still believe the latter is achievable.

Australia and the USA are perhaps the only places around the world where the illusion of sufficient road capacity for free-flowing traffic still persists. Perhaps the enticing hints of it that you see at night or early morning are what provides the illusion that it might still be achievable. It’s not.

Most people prefer to drive? Well of course they do!

“But people prefer to drive!”

Well of course they do.

Investment since 1930 has been heavily skewed towards roads. Just look at how many railway stations opened between 1930 and 2010.

Most services are infrequent. Even on the train network, many lines run only every 20 to 40 minutes on Sundays, for instance.

Buses are even worse. Mostly hourly on weekends, and only every half-hour on weekdays – even in peak.

With so few frequent services, they should be co-ordinating timetables so things connect. But they’re mostly not, or at least, progress is slow. (And even if the times line up, whether connections wait if delays occur is unclear. Even Metro’s own shuttle services to Alamein and Cranbourne won’t usually wait for late connections… to do so might caught the delays to cascade in such a way that the service won’t recover.)

In fact, have you noticed how the train network (including V/Line) is moving towards a 10/20/40 minute frequency, but most buses are every 30/60 minutes? No wonder the timetables aren’t synched!

We’ve got the bones of a good network. The infrastructure isn’t perfect, but there’s plenty of capacity outside peak times: track capacity, fleet, bus bays/stops/zones.

August 2018 timetable changes

The opening of the rail extension to Mernda triggered a range of timetable changes on trains and buses around Melbourne.

This follows on from train timetable changes last year on the Craigieburn and Sunbury lines, and tram timetable changes earlier this year.

PTV has a page summarising the changes which is worth a look.

Dandenong line evenings

The main change was extra evening services. Unfortunately this State government press release from 4/6/2018 over-reached:

Dandenong Corridor passengers will benefit from up to 80 new and extended services each week and a longer turn-up-and-go service frequency in the evenings, meaning trains every 10 minutes until 10pm on weeknights.

Most people would take that to mean that the trains are every 10 minutes until 10pm from the city. In fact it’s until 9:33pm, then drop to every 20 minutes until about 10:30pm, then every half-hour.

Undeniably it’s progress — previously the 10 minute service finished at 8:22pm. But it’s not quite what was implied.

Newport lines

An extension of frequent services between the city and Newport better matches the promise:

Thirty-five new weekday services will also be added on the Werribee Line, and passengers at South Kensington, Seddon, Yarraville and Spotswood can catch a turn up and go service until 8.20pm on weekdays.

Bus connections

The principle of changing bus timetables when train timetables change is good one. In some areas, work has been done to align connections. This helps explain why some bus routes run every 20 or 40 minutes, in areas with trains every 20 minutes.

PTVFrom Sunday 26 August, a number of bus timetables will change to keep them connected with new train timetables on the Werribee, Hurstbridge, Cranbourne, Pakenham and South Morang/Mernda lines.

I noticed one of the routes that’s changing is local to me, the 701. So I took a look at the new timetable.

Saturday night 701 buses depart Oakleigh: 7:05, 8:11, 9:11. This appears to align with trains arriving 6:56, 8:06, 9:06.

Sun night 7:16, 8:16, 9:16, trains arrive 7:06, 8:06, 9:06. Seems okay (though connections at the other end of the route at Bentleigh are less good).

The great Bentleigh crime wave

I started writing this during the state election campaign. The Coalition campaigned strongly on crime, and even claimed Labor had cut police numbers in some areas, including Glen Eira. Labor claimed the opposite.

Is there actually a crime wave? Are home invasions up?

The Crime Statistics Agency says that overall crime was falling from 2008-2011, then headed upwards, peaking in 2016, but it’s dropped again in 2017 — hopefully the start of an ongoing trend.

But Residential Aggravated Burglary is way up — almost doubling from 2008 to 2017. Local area figures for Glen Eira show this trend as well.

Victoria - crime statistics 2008-2017

So there is some truth that some types of crime are increasing, while the overall crime rate seems to be dropping.

The problem with the stats vs the rhetoric is it’s often a bit hard to pin something like this directly onto either side of politics. And not very helpful of course.

Bear in mind that the rate of recorded criminal incidents could go up as a result of better reporting. For instance we know that recorded offences rose over the period that Protective Service Officers were deployed onto railway stations. I’m not sure that would be the case for aggravated burglary though.

What about police numbers? Are they going up, or being cut? The claims above can’t both be right.

Victoria Police employee numbers for June 2018 show a total of 14,476 Police Officers (Full-Time Equivalents) — this excludes 1,365 PSOs. In Glen Eira there are 93 police plus 8 public servants = 101 total.

The December 2014 figures showed a total of 13,135 Police Officers. In Glen Eira: 98 total (no breakdown). So to me, it looks like both are up, though not by very much in Glen Eira. And it doesn’t seem unreasonable to deploy most new officers to other, growing suburbs.

In my experience, politicians don’t usually lie. But they often do selectively cherry-pick statistics, and bend the truth. Usually somewhere in the stats, there’s some basis, however tenuous, for a claim.

Is fear-mongering on crime even a good political strategy in a relatively safe area?

It might work in the areas that actually have big issues with crime, but I know some locals in my area who are entirely unconvinced by this.

I think my hunch proved right. The Vic Coalition’s 2018 election strategy was a flop.

NE Link and the bus way

North East Link’s latest costing is now an eye-watering $16 billion.

And a bus way thrown in, to provide fully-segregated bus lanes along the Eastern Freeway. This would bring some benefits… but… any busway on the Eastern Fwy:

  • puts at risk future rail
  • could be done without spending $16b on NELink
  • doesn’t address where most Doncaster bus delays occur – between Clifton Hill and the City

Airport rail link

November 2017: Airport rail plan starting to move.

It’ll still run via Sunshine/Albion, but importantly there’s been a change of strategy: the latest thinking is making it part of a wider network upgrade that includes more separation of regional and suburban lines — the latter in the west is facing huge patronage growth due to urban development.

And the idea is that the airport service won’t be a dedicated line with no other uses. This is good — as I noted when visiting London recently, specialised point-to-point express trains tend to be expensive and not used by many people.

Mind you, have you noticed how both big announcements by politicians about airport rail recently are both completely different from PTV’s “long-term” Network Development Plan?

  • PTV NDP (2012): branch at Albion from the Sunbury (to Cranbourne/Pakenham via tunnel) line
  • The Coalition’s plan before the 2014 election: branch at Albion from the Sunbury (to Cranbourne/Pakenham via Flinders Street viaduct 
  • Labor’s plan now – apparently some kind of fast rail line via Albion, possibly as part of an outer-suburban connection to Wallan and Clarkefield, alongside upgrades to services to separate Melton and Wyndham Vale services from V/Line trains to Geelong and Ballarat

The details are very vague at the moment. It’ll be good to see this plan fleshed out a bit.

“The CBD is at a standstill”

When you hear traffic/media reports that the CBD is at a standstill…

The reality is that those in the traffic are at a standstill. Sure, it’s very visible. But if handled well, it doesn’t actually affect the majority of people, because most people don’t drive into the CBD.

Pedestrians can face congestion, but can mostly keep going – if they’re able to walk around the illegally stopped cars on the crossings.

Trams often have some problems while diversions were worked out. Some bus routes can be affected by the rapid spread of car congestion, though some that are able to use tram lanes (such as across Queensbridge) can keep moving.

Trains keep running. In one case I’ve seen reports of cross-CBD car trips taking up to an hour. That’s double the time it takes me by train from Flagstaff to Bentleigh (which of course includes crossing most of the CBD, underground).

Station incident, November 2017

You may have seen this footage:

One report said the police was not naming the location, but… well, it’s very clearly Windsor.

After the initial reports from channel 7, picked it up.

Similar stories appeared in the Irish Independent and the London Evening Standard.

I will just note that some of the reporting got a bit muddled.

Australia has seen many incidents similar to this in recent times and nobody yet has died as a result, according to Public Transport Users Association spokesman Daniel Bowen.

“Platform screen doors are difficult to fit retrospectively, which is why they are being fitted on new lines like the Melbourne Metro for example, but not on existing stations,” he told Channel 7.

What I actually said was that there had been incidents of prams falling onto tracks, but no deaths. I’m not actually sure if there have been fatalities due to incidents like this one.

Oh, and all those comments were made to, not Channel 7.

Ads on public transport

How much money is handed over to provide ads on public transport vehicles and shelters? Quite a bit, though most pricing is not readily available in public.

Here’s on deal from 2017 – it seems to refer to a bus stop shelter ad: $168/week +GST (minimum 34 weeks) – that’s about $9500 per year.

Another price I found a while ago was ads on the back of a bus: $521 per week.

No wonder operators love that supremely annoying all-over-advertising. It must bring in a fortune.

Happy new year everyone!

Photos from ten years ago Toxic Custard newsletter

Old photos from December 2008

My regular post of ten year old photos

The smiley in the sky (first posted here)
Smiley in the sky

Tram of the beast (First posted here)
Tram of the Beast

The next station is Connex
The next station is Connex

Back when the trams were battleship grey, and they used obscure route numbers like 47.
Tram in Collins Street, December 2008

In December 2008 a brand new bike parking cage was under construction at Bentleigh. It would be demolished just 7 years later as part of the rebuilding of the station.
New Parkiteer cage under construction at Bentleigh, December 2008. (Demolished: 2016)

Selfie on a train
Selfie on train, December 2008

I’m not even sure why I snapped this photo: view of the city from above the Bunbury Street tunnel in Footscray. It certainly wasn’t high resolution by 2018 standards, but you can view it larger here.
Bunbury Street tunnel, looking towards the City, December 2008


Boxing Day buses at Chadstone: still big problems

Last year I wrote about Chadstone’s Boxing Day bus debacle: delays, heavy demand, traffic jams, queues, and not a single additional bus deployed.

So, how was it this year? Spoiler: Almost the same.

Here’s some video, or read on.

So I wanted to head to Chadstone to look for a bargain. It’d be crazy to drive and get stuck in traffic, right? Let’s do the right thing and take public transport.

I boarded a 624 bus at Murrumbeena to go to Chadstone. Just like last year, it was slightly delayed (4 minutes) and packed.

624 bus to Chadstone, Boxing Day 2018

The traffic on approach to the centre was very congested. Poath Road was busy, and so was Dandenong Road eastbound, with a traffic snarl at the southern entrance into Chadstone – no thanks to numerous drivers ignoring our old favourite rule 128 and blocking the intersection trying to get in. Our bus driver had to take the rightmost lane (of three) to get around it.

Traffic at Chadstone, Boxing Day 2018

VicRoads had said they were adjusting traffic light sequences around the centre. A little enforcement of the rules would have helped a lot.

Dandenong Road buses were diverting “around the block” to Warrigal Road, where Chadstone management put a lot of effort into allocating a bus lane from into the bus interchange. Once we reached it, this worked quite well – kudos to them for this.

Bus lane from Warrigal Road to bus interchange, Chadstone, Boxing Day 2018

However the traffic lights to get in and out of the bus interchange need attention. It was a long wait to get in, and buses exiting again also had a long wait. We finally arrived 18 minutes late.

Some routes had to weave their way in, out again and back in to get to the correct stop in the interchange, making already delayed buses worse.

Buses queued at Chadstone to exit the bus interchange, Boxing Day 2018

The crowds and queues of passengers at the interchange seemed just as bad as last year. This time, police were patrolling (not that they had much to do), and PTV and/or bus operator staff were helping direct passengers, as some of the usual stops were changed around.

Buses at Chadstone, Boxing Day 2018

Delayed buses caused flow-on effects right along the routes serving Chadstone. This 900 bus from Rowville and Chadstone arrived at Caulfield more than 20 minutes late to run its next service back, leaving passengers behind.

Queue for bus at Caulfield to Chadstone, Boxing Day 2018

My conclusions from observing all of this

The traffic lights in and out of the bus interchange need adjustment. Making buses wait extended periods just adds to the delays.

The temporary bus lane worked well. Bus lanes from Warrigal Road to and from the interchange probably ought to be permanent. It’s 6 lanes (3 in, 3 out, expanding to 5 out at the intersection). It’s also a crying shame bus lanes weren’t provided from Dandenong Road the short distance to the bus interchange, when it was built.

Boxing Day sales are a known event. They happen every year. Everyone knows that huge numbers of people will turn up. We know lots of people will try to catch buses.

We know the traffic will be awful. Extra people in buses is fewer people in cars. This is a good thing, for the centre and for other people who do drive.

Boxing Day cricket attracts around 100,000 patrons. For this, they deployed (by my count) 27 extra train services, on top of the 10-20 minute standard public holiday frequencies.

Boxing Day at Chadstone attracts at least 120,000 patrons. For this, despite crowding year after year, they deployed not a single additional bus. Zero extra services. Smartbus routes mostly every half-hour, other routes mostly hourly, with some severe delays of 20+ minutes on top of that. It’s completely inadequate.

As I said last year: this is a planned special event for Chadstone. Is it treated as such under the Transport Act (see sections 192 to 204)? This can mandate traffic management and can require additional services funded by the event organiser – it’s very obvious that this is precisely what’s needed.

This problem is not confined to Chadstone, and not just on Boxing Day. Essendon DFO had all its route 903 buses divert away due to traffic concerns. Other suburban malls would have had similar issues, and bus crowding is a problem every weekend due to infrequent services.

There are plenty people willing and able to catch buses to major suburban shopping centres. They’re doing the right thing. It’s time the buses were fixed.

PS. Some off-piste responses to this issue on Twitter included:

  • Bus drivers deserve time off. But hundreds of them worked on Christmas Day and Boxing Day. The additional staff needed to run some extra services to hot spots like Chadstone is minimal in comparison.
  • Bus passengers don’t spend as much, so shopping centres don’t care. Okay it’s true car passengers might fill their boot with stuff, but how many of them actually do? Specifically, how many return to their car multiple times before driving off? Shopping centres know they have finite parking capacity. Every person who arrives by another mode is an additional shopper. And as the photos show, plenty of people were willing to use the buses. Even more would if they provided a decent service.
  • You could have shopped on another day! Really? So me personally not going to Chadstone on Boxing Day would have fixed the problem?

Another PS: the scene at 10pm:


Melbourne’s public transport patronage since 1947

One final blog before Christmas.

PTV published this data some time ago, covering Melbourne’s train/tram/bus patronage from 1947 to 2011, but I’ve added the most recent figures using numbers from the Budget Papers.

The data starts just after WW2, with the highest number of trips shown being 587 million in 1948-49.

Melbourne public transport: millions of trips per year

There was a big dip for trams in 1949-1950, and another for trains in 1951 due to extended industrial action. The PTV report also notes that post-war petrol rationing finally ended in 1950.

It’s notable that the early figures show more tram trips than train, perhaps reflecting that until that point, Melbourne’s suburbs were mostly still within reach of the tram system.

Trams also play a continuing role for inner-suburban destinations, including for trips partly served by train. There were still more tram trips than train most years until early this century, when train patronage started to accelerate fast.

Across all modes from the 1950s, there was a sustained downward trend until the 1980s. The public transport network wasn’t expanding – in fact it was contracting, with most Footscray-area tram routes closing, and other services being reduced. Cars were getting cheaper, and governments were spending up big on roads, encouraging more and more people to drive instead.

The low point is 1980, with just 90 million train journeys, and 262 million trips on the network overall. From there, with the opening of the City Loop in 1981, and the introduction of multi-modal tickets also in 1981, things begin to pick up a bit.

There wasn’t much change through the 1990s – but by the end of the decade, moderate increases are seen. I’d tend to attribute this to recovery from the 1990s recession, and big increases in Sunday tram and train services in 1999 (suddenly it became practical to use public transport on any day of the week), and the continued roll-out of new (air-conditioned) trains and trams.

PTV-liveried train, tram and bus

The virtuous circle

It’s in the mid-2000s that Melbourne’s population starts to take off, alongside CBD growth, and for trams and trains (not so much buses) we’ve got into a “virtuous circle”: extra patronage leads to crowding. Crowding prompts action and investment, which results in service upgrades, which ultimately makes the system more attractive, leading to extra patronage. Rinse and repeat.

Even relatively small things like running 6-car trains right through the week (instead of 3-cars in evenings and on weekends) helped remove a barrier to patronage growth. Better information including apps, and signage is a help as well.

Helping all this along, I think, was an increasing awareness of the importance of public transport in a big city. It’s also worth noting that in the mid-2000s, the headquarters of the two leading television networks (7 and 9) moved into the CBD, and the trend in recent years has been for more journalists to use the public transport network themselves, helping to bring issues such as crowding into focus.

Where the virtuous circle has been less apparent is on the bus network. The big upgrades in recent years were the introduction of the Smartbus routes, in stages mostly between 2005 and 2010, and university shuttles (especially the Melbourne Uni 401 in 2007, the Monash Uni 601 in 2011), with some local network upgrades in areas like Wyndham, Brimbank and Mernda.

All these have resulted in more patronage. The problem is that regular crowding has not prompted a very strong government response.

In 2008, train patronage surpassed the 1950 peak. The overall network total hasn’t yet beaten that peak, nor have tram or bus patronage figures, though they are on the way.

The numbers don’t show it, but I would expect some key differences between 1950 and now are:

  • Fewer people had cars back then to cater for non-work and non-CBD commutes, so trips now tend to be more concentrated on work commutes these days. This means bigger fleets and more track capacity are required now, because lots of people are travelling at once into major destinations like the CBD.
  • The spreading of Melbourne means trips tend to be longer, especially train trips – helped along by the 2015 introduction of largely flat fares
  • That said, in inner suburban Melbourne, increasing numbers of people are choosing to live without a car, meaning non-work trips are on the rise again, helped by the completion of projects like Southland station.

Overall, public transport usage is increasing. And while there might be growing pains as investment catches up, ultimately, it’s got to be good news.

Merry Christmas!

PS: There will be a follow up post that looks at population growth over this time.


A quick skim of the Annual reports

State government annual reports season ended up hitting late this year due to the election.

It’s usually in September, but this year everything got tabled in Parliament late on Wednesday.

I usually like to browse through the V/Line and PTV reports for interesting factoids…

Here are some things I noticed during a quick skim.

V/Line annual report 2017-2018

20.8 million trips, up about 1.5 million in a year. No wonder the trains are crowded. All of the increase is on rail; coach trips are slightly down to 1.3 million, and account for only about 6% of all trips.

They sold 2 million paper tickets (used for coaches and trains outside the Myki area), so that might mean that around 90% of V/Line trips are on Myki, though some paper tickets would presumably cover multiple (eg return) trips.

Compensation paid doubled to $157K – and that’s “compulsory compensation”, so presumably not voluntarily, but related to ongoing (worsening?) poor performance.

Farebox is up to $102m, but full fares are down 5% to 62%. The subsidy per passenger increased slightly to $23.93

V/Locity carriages increased by 27 as locomotives decreased by 10 – apart from newer trains being hopefully more reliable, the move to Diesel Multiple Units (DMUs) over time should help cut costs, as there’s none of the messy shunting the loco when terminating each service

Geelong line patronage up an amazing 115% in the past 4 years

Trespasser incidents up from 589 to 908, but this is attributed to better reporting rather than increased occurrences

It’s interesting to read the summary of staff numbers. Out of 1915 Full-Time Equivalents), the frontliners are: 301 station staff, 289 conductors, 379 train drivers, and just 9 Authorised Officers

V/Line trains at Southern Cross

PTV annual report 2017-2018

Metropolitan trips 565 million for 4.8 million population… I make that 117.7 trips per person per year

Regional trips 33.3 million for 1.5 million = 22.2 trips per person per year; perhaps no surprise that it’s lower, but it makes me think a lot more needs to be done to improve regional town services, where public transport can compete more against car trips.

Metropolitan services fare recovery 30%

… but total fare receipts are on the rise: 2018 $917m, 2017 $871m, 2016 $814m, 2015 $845m, 2014 $806m, 2013 $604m. (Next time some bright spark suggests making public transport free, ask them where the $917m in revenue will come from instead if there are no fares.)

Payments to operators of $1.4b to MTM (Metro), $0.3b to KDR (Yarra Trams), $2b for the dodgy state government Capital Assets Charge for rail infrastructure (there’s no equivalent road charge) and $1b for bus services.

(As I understand it, MTM and KDR also get 40% each of the metropolitan fare revenue, so it appears to mean the trams come close to breaking even, despite all the Free Tram Zone free rides.)

Nothing in the reports leapt out at me as a gotcha reveal, but things are so busy that I’ve only had time to skim.

And among the other 115 reports released on Wednesday, the VicTrack report is also out… I haven’t read that yet.

Anybody else spot anything interesting?


Bus punctuality stats finally public… well, some of them

The Age’s Timna Jacks got hold of some bus punctuality figures via FOI, and published this story last week, which is worth a read:

Late every second trip: Life on the delayed 232 bus

The data provides a glimpse of something not normally in the public eye. Regular bus users know that some routes suffer greatly from lateness, but the statistics aren’t usually published.

With Timna’s permission I’ve dug around in the data a bit more to see what other insights can be found.

What’s in the data: Punctuality to within 5 minutes for Transdev routes from 2013 to 2018. What it doesn’t show is service delivery, ie cancellations and short run services, which also influence how reliable and usable bus services can be.

There are also operator-wide average figures for Ventura and Dynson from 2015 to 2018.

Towing a bus

Transdev’s best and worst performers

The data had a route-by-route breakdown of Transdev’s punctuality since September 2013, from shortly after they started operations after winning their contract and taking over routes formerly run by Ventura and Melbourne Bus Link.

The five worst Transdev routes (out of 47), based on average punctuality from 2013 to 2018:

  1. 220 Gardenvale – Sunshine (via City) 60.18% punctuality, but “winner” of the worst single punctuality figure, with just 38.7% of buses on time in August 2015
  2. 216 Brighton Beach – Sunshine Station (via City) 64.89%
  3. 219 Gardenvale – Sunshine South (via City) 65.58%
  4. 232 City (Queen Victoria Market) – Altona North 65.75%
  5. 350 La Trobe University – City (Queen St) 68.77%

The worst three are the long cross-suburban routes, running via the CBD, with almost no bus priority measures along the route. Just in the past few months (after the data ends), these have been split into two, theoretically temporarily, to avoid even worse problems while the metro tunnel is being built… so punctuality is likely to be improving now.

Rounding out the worst five are the 232, which improved in 2014 when the Port Melbourne diversion was removed – and route 350. Both these routes also serve the CBD. In fact 8 out of the 10 worst routes serve the CBD – no surprise perhaps, since that’s where traffic is likely to be at its worst.

(Can’t see the graph? Click here)

Some of Transdev’s routes benefited from timetable changes in 2014, and there was a more thorough change in 2016.

These changes mean there has been improvement, but looking at just the worst average punctuality for 2018 (January to June), the same five routes feature, with route 232 is now the worst, with 64.23% of buses on time.

What Transdev wasn’t able to do was introduce its “greenfields” timetable, which would have made wide-ranging changes, including splitting the long orbital Smartbuses (and some of the cross-city routes) up into more manageable, logical routes.

That proposal would have overcome a lot of the punctuality and crowding issues, but it also would have cut services in some areas, because it was basically a reshuffle of existing resources, where ultimately, more are needed. Labor rejected the plan when they came into office in late-2014.

Tram/bus stop, Queensbridge Street, Casino East

And the most punctual Transdev routes? Unsurprisingly they’re all in the middle and outer suburbs, and mostly relatively short routes of under 10 km.

  1. 284 Box Hill – Doncaster Park & Ride 89.55%
  2. 370 Mitcham – Ringwood 88.21%
  3. 295 Doncaster SC – The Pines SC 87.69%
  4. 279 Box Hill – Doncaster SC/Templestowe 86.42%
  5. 380 Ringwood Loop 85.62% – this is the only route of the top 5 to be over 10km long

Perhaps surprisingly, the “DART” City to Doncaster Smartbuses 905, 906, 907 and 908 all rank in the top third of routes for punctuality. They do have bus priority measures along much of their routes in peak hour, though there are problems with motorists invading the bus lanes on Lonsdale and Hoddle Streets, and little or no enforcement. (And remember, the stats don’t measure cancellations.)

Somewhere in the middle of the pack are the long orbital Smartbus routes 901, 902 and 903 – they do better than one might expect because they have plenty of fat in their timetables at various points along the route.

All of the Smartbuses benefit from long operating hours and relatively frequent off-peak services, which can help bring up the average punctuality compared to routes that run most or all of their services in peak hour.


Comparing bus operators

The data also provided operator-wide figures for Ventura and Dyson. Here’s a comparison for those two plus Transdev*, with Yarra Trams thrown in as well:

(Can’t see the graph? Click here)

As you can see, punctuality generally improves for a month in January, as traffic (and patronage) is lighter than usual, despite frequent construction blitzes.

Ventura consistently has fewer buses running late than Transdev or Dysons, but it’s worth noting that Ventura run routes predominantly in the middle and outer suburbs, with none of the inner-city or CBD services which tend to suffer worst from traffic congestion.

Transdev was the worst of the three during 2015, but lifted its game in mid-2016 with timetable changes across most routes. Since then they have on average seen fewer delays than Dysons most months.

So on this measure at least, Transdev has been improving.

Remember: this is punctuality only. The figures don’t measure cancellations, short running, cleanliness, buses catching fire or taken off the road by the safety regulator, or any of those other things for which Transdev Melbourne is notorious.

What more can be done?

On Saturday the State Government announced that 100 of Transdev’s fleet (which is owned by the government) would be replaced by new buses. This should help reliability (provided the operator actually maintains them properly), but may not help with punctuality and other issues such as crowding.

It was also announced that Transdev’s contract would end in 2021 — only one year of the 3 year option is being exercised.

Upgrading buses is a good idea. Hopefully it’ll bring better reliability. Better passenger information would be welcome. Perhaps they should use the opportunity to start a transition to electric buses, particularly for inner-city routes.

But beyond that, it’s important to take other measures to improve bus services. This could include:

Proper bus priority, with more bus lanes and traffic priority, and enforcement of those measures

More frequent services to cut waiting times and relieve crowding, which is chronic on some routes, and adds to delays

Bus route reform to make more efficient use of fleets and drivers, and make the network easier to understand and cut overall travel times by making routes more direct

Contracts that are structured so appropriate funding is available to properly maintain the fleet, and run the service to a decent standard – this appears to be where the Transdev Melbourne contract has gone wrong

Public performance data, so there’s more transparency. I don’t think it’s a complete coincidence that trams and trains have (some) public performance data, and get the bulk of the attention and investment

And yes, timetable adjustments where warranted – but this should be the last resort, not (as now) the default action when buses are regularly running late. Adjusting the timetable just means making those delays permanent, not actually fixing them.

Buses serve many parts of Melbourne that don’t have trams and trains, and may never get them.

It’s high time the government treated them seriously.

  • Note: The overall figures for Transdev have been calculated by using the average of the route punctuality figures. This is not the same as the average across all Transdev services, as some routes have a lot more services than others.
  • See all the data here. Got further observations? Leave a comment!
  • Marcus Wong also has a post on Transdev today, focusing on recent problems with bus reliability and maintenance

Fares for short trips now double the cost of Sydney

As expected, public transport fares in Victoria go up from 1st January. This time it’s a CPI rise of 2.2% – thankfully not as high as the last four which were CPI+2.5% (budgeted by the Coalition in 2014, delivered by Labor).

This takes the standard zone 1 or 1+2 daily fare to $8.80, and the 2-hour fare to $4.40.

It’s worth noting that for short trips in zone 1, this is now precisely double the $2.20 Sydney tram or bus Opal fare for up to 3 kilometres.

The fare rules are a bit different – Sydney’s free transfers are generally limited to the same mode within 60 minutes, and the daily cap is much higher (except on Sundays).

(The short distance single zone Go Card fare in Brisbane is $3.25 peak, $2.60 off-peak. In Perth the Smartrider fare is $1.98 for two “sections”, or $2.79 for one zone.)

It turns out that the fares in Sydney in most cases are lower than Melbourne’s standard $4.40.

  • Sydney tram/bus: $2.20 for up to 3km, half the Melbourne zone 1 cost, and 26% cheaper than Melbourne zone 2 at $3
  • Sydney tram/bus: $3.66 for 3-8km, still 17% cheaper than Melbourne zone 1, though more expensive than Melbourne zone 2
  • Sydney train: $3.54 peak/$2.47 off-peak for up to 10km
  • Sydney train: $3.08 off-peak for trips of 10-20km. (The peak fare is identical to Melbourne, at $4.40)
  • Even train trips of 20-35km off-peak are cheaper in Sydney, at $3.53

Melbourne’s fares (within zones 1 and 2) are generally cheaper for longer trips: over 8km on buses (Sydney trams don’t even go that far!) or over 20km on trains in peak, or over 35km off-peak.

And the transfer rules and daily cap mean Melbourne is cheaper for roaming around all day (for instance, tourists) – but this is not what most daily commuters do.

It’s those short trips that really sting, thanks to the nearly flat fare structure. You wonder how many people choose the car by default for short inner-city trips because of it, especially when travelling in a group.

Also spare a thought for V/Line passengers. Anomalies in the zone system mean that, for instance, a passenger from Lara to Melbourne pays $4.40, but hopping on the train a few minutes earlier at Geelong will cost a whopping $13.40 in peak, $9.38 off-peak.

How much have fares gone up over time?

Looking at the basic 2-hour zone 1 adult fare, it’s gone up quite a bit over time – from $1.60 in 1990 to $4.40 in 2019. If it had followed the rate of inflation, it’d now be about $3.57.

Melbourne PT fares since 1990

See the data. I am currently missing the prices for 1993. Can anybody help?

People travelling long trips (the old zone 1-2-3) have benefited from the removal of zone 3 in 2007, and the fare capping added in 2015. In 1990 the 2-hour fare was $3.80, from January it’ll be $4.40. If that had followed the rate of inflation, it’d be about $8.49.

There was a slight drop for all fares in 2013 (actually in the last days of 2012) when single use Metcards were phased out, forcing everyone onto the bulk price Myki fares.

Solutions are hard

Solutions are hard because so often the politicians reject any policy change where anybody is disadvantaged.

The flat fare has pros and cons. The only logical alternatives are more zones, or point-to-point pricing. The latter is difficult given the older Myki readers still common around the system on trams and buses are so unreliable, and Myki zone detection is so hopeless.

It appears for instance that buses have two GPS systems. One is able to, usually, tell the real-time apps how far away the bus is to a minute’s accuracy; the other is often unable to figure out which of two gigantic zones it’s in.

Introducing off-peak fares is another option. V/Line gives a 30% discount for off-peak. Perhaps this should be looked at for metropolitan trips too. The same discount would bring that $4.40 fare down to $3.08, matching Sydney on more trips. This would help improve demand at off-peak times, and would shift some trips out of peak.

There is a kind of off-peak pricing: between 6pm and 3am you’ll only pay one 2-hour fare – this was inherited from Metcard, and inherited from the 1980s era paper tickets before that! There is also the weekend/public holiday $6.30 cap.

And another upgrade that would be of benefit: re-instating the weekly and monthly caps once planned for Myki would help stop Myki Money users paying too much, and remove the confusion of having to choose between Money and Pass to get the best deal.

Toxic Custard newsletter transport

Desire lines: signs of bad design?

Desire lines are where authorities intend for people to go one way, but people (especially pedestrians) quite logically ignore them and go a different way.

Often they indicate poor design.

Here are some quick examples from my neck of the woods.

You have to wonder whose bright idea this was. Try and divert the pedestrians away to a crossing. Why do it? The worn grass indicates not many people follow the recommended path.
Desire lines in Bentleigh

Similar story at this roundabout. It’s a less busy street for pedestrians so the grass looks more intact, but again, why? Puzzling since another roundabout 100 metres away doesn’t have this design.
Desire lines in Bentleigh

Down at Southland, the new station is a roaring success… except for the pathway to the shopping centre, which diverts people via an indirect route – though at least it’s got priority zebra crossings all the way – visible at the left. Still, an awful lot of people come out of the station and instead dodge around the fence and make straight across the car park for the entrance. Are we really that surprised? Hopefully sooner rather than later, Westfield will fix it.
Southland station desire lines

The centre of central Bentleigh: the station. This new pedestrian crossing is very welcome, as it connects the westbound bus stop with the trains. Amazingly, before the grade separation, there was no nearby crossing. With a little thought, they could have made this new crossing wider, stretching towards the bus stop, as when buses arrive, there’s a swarm of people crossing the road.
Bentleigh station pedestrian crossing

And this, around the corner. Having a zebra crossing is good, but it’s clearly in the wrong place. It should be no surprise at all that most people cross at the point aligned with the supermarket entrance. Authorities must have realised this, or there wouldn’t be this signage.
Desire lines in Bentleigh

Often this type of thing appears to be just trying to make life difficult for pedestrians.

I really hope whoever is responsible for these designs is observing how people use these spaces, and isn’t continuing to make these mistakes.

More reading: Desire paths: the illicit trails that defy the urban planners


Renaming North Melbourne

It’s perhaps a minor thing, but…

Back in November 2017 the metro tunnel station names were announced, including a new station called North Melbourne, with the current station to be renamed to West Melbourne.

I don’t have a problem with that. It’s logical… but it has to happen well in advance of the new station opening, to minimise confusion.

At the time, there were assurances that this would be done quickly.

A good opportunity to rename a station is when there are other rail network changes, prompting a new edition of the rail map.

Publishing a new map is a big thing. It has to be designed, and then every map around the system needs to be replaced. Web sites are easy, but there’s every station, every train carriage.

Well they missed an opportunity. In August 2018 the Mernda rail extension opened, prompting a map revision. But “West Melbourne” is still “North Melbourne”.

PTV train map August 2018

If you want an illustration that replacing all the maps is a big task, note that many stations, including my local, are still displaying the old pre-Mernda maps.

So, what are the next metro stations to be opened? Well there’s actually nothing definite until the tunnel opens in 2025… including the new North Melbourne.

They could have done the renaming in August, but they didn’t.

Perhaps there are other factors holding them back; some behind-the-scenes issues that need to be resolved. Presumably those can be overcome; just as they were for Melbourne Central (Museum) and Southern Cross (Spencer Street).

There is another chance before 2025. There aren’t any more metro stations expected before then, but there is a V/Line station: Cobblebank, in Melbourne’s outer west, slated for completion during 2019.

Let’s hope they take this opportunity to rename North Melbourne well before the new station opens.