This video on the PTUA feed yesterday was mostly compiled from stuff I filmed in Barcelona last year:
It is obviously not intended to be a direct comparison between Barcelona and Melbourne.
They’re different cities, though Melbourne and Barcelona have very similar metropolitan populations (5.44m vs 5.79m).
Barcelona is a bit less than half the geographical size (9993 square km vs 4268), meaning the population density is about double there.
How many lines?
When comparing rail networks it’s a bit complicated. Melbourne’s got 16 suburban “Metro” train lines (excluding Racecourse) but you could argue it’s really 17, if you don’t count Stony Point (a very infrequent line through a mostly rural setting), but do count the V/Line Wyndham Vale and Melton lines (busy suburban lines, even if run by diesel trains).
Barcelona is 12 Metro lines, and they also have suburban lines. But even this is wobbly, as metro L6 doesn’t run in peak hour, making way for frequent suburban services.
Patronage
Leaving those niggly details aside, Melbourne’s Metro rail patronage (eg excluding V/Line routes) was 189 million for the year ended March 2026.
Barcelona’s Metro (excluding suburban lines) was about 440 million in 2023. (I can’t immediately see any later data, and it’s not crystal clear to me if this includes just TMB, or also the two lines operated by FGC.)
Barcelona’s Metro definitely has higher frequency, with trains every few minutes all day, which encourages higher patronage. Partly this is necessitated by trains about half the length of Melbourne, but even so I found the frequency to be more than double that of most Melbourne lines.
But as I say, the video was not intended to be a direct comparison. This earlier video from Paris spelt this out:
We are not those other cities.
But we can learn from them, and be more like them. Maybe it’s impractical for our trains to run every 2-3 minutes all day, but no way should passengers be regularly left waiting 20, 30, 40 minutes to get moving.
6 replies on “BCN vs MEL”
I’m curious to know what level of costing a modelling has been done around increasing service frequency.
In particular:
– The cost of 20, 15, 10, 5 minute frequencies.
– The point at which each line / group of lines becomes saturated using existing track infrastructure.
– The point at which fleet sizes constrain the frequency.
– Expected passenger numbers for each service frequency, with expected modal shift from private transport.
I appreciate there would be vast amounts of data and assumptions within this, and would value it being condensed down to a single matrix with RAG status – rail line vs frequency.
@Dan, the main costing I’m aware of is around 10 minute trains. The PBO reckoned $173m per year, which is less than half the cost of the free and half price fares this year ($432m).
https://danielbowen.com/2019/10/23/how-much-would-trains-every-ten-minutes-cost/
As for 20 minutes, thankfully that’s mostly happening anyway.
https://danielbowen.com/2026/04/30/progress-on-train-frequencies/
“the main costing Iām aware of is around 10 minute trains. The PBO reckoned $173m per year, which is less than half the cost of the free and half price fares this year ($432m)”
Geez that seems a pittance in the scheme of the mega billions spent the past 12 years. I’ve asked my local MP about when service will improve and why not already- text book answers again that show the government seems only focused on infrastructure (although needed) but not boosting service much when it could have begun some time ago and certainly in the past 12 months. That it isn’t even in the budget for every line, bus and tram shows how out of touch things have become with Govt/DoT (or whichever dept it is now running things)
Encouraging higher patronage is not just a matter of more frequent trains, but also how easy it is to get to/from the station in the first place. So, not just “how many lines” but also “how many stations” (or rather “what station density”). Wikipedia says the BCM has 189 stations while MEL has 227; you say that Barcelona is a bit less than half the size of Melbourne, so that’s about double the station density. Nice.
Also, regarding the half-size trains, that would only necessitate double the frequency if they were running to passenger capacity, which they are nowhere near doing during the day. Off-peak daytime trains that I observe are probably around 10% occupied if that.
@Daniel – Thanks for the link back to the 2019 article. Given the cost relative to other initiatives, I’m amazed there hasn’t been a government which was prepared to trial it for a year or two. It seems a no-brainer. š¤Æ
The problem with more services from the government perspective is probably this:
– Raising service levels requires more drivers
– Once you hire the drivers, you probably don’t want to fire anyone after putting all the work into training them.
The government absolutely despises higher opex, is my supposition. Never mind the financial illiteracy in spending 35B on SRL but not 200m a year on services.
It is a complete joke that in the context of all our infrastructure projects the government can’t find the relative pittance it would cost to run high frequency services round the clock. 5B set aside could fund perpetual high service levels forever on interest alone – and that’s 1/7th the cost of SRL, and is in any case likely a prerequisite to realise real benefits out of something like SRL.
It’s even sillier when you consider that there are probably more staff watching youtube in those little offices across the network than there are drivers sitting in trains – to be frank, any private business would likely have reallocated most of the money used to pay them to higher service levels instead.
@Joe Green – More trains makes train stations easier to get to, weirdly enough. If you catch a bus to the train station, knowing that no matter what you only have to wait max 10 minutes to catch the train is a massive benefit. Not having to think about whether you’re going to miss a train and be forced to wait half an hour is a big draw for more patronage.