Discussion from CRR thread on commuting

[SteelPan]

"Cross River Rail will completely transform the way Queenslanders and visitors will travel....."   

[Jonno]

Works-class, innovative, leading-edge, superdooper amazing "worst-practice" public and active transport mode share!!

[#Metro]

We got the tunnel.
Let's focus on 15 min services within the 30 minute commuter zone and all day express services outside of that.

[aldonius]

Commute zone's a lot bigger than 30 minutes haha.

But yes. The concrete's locked in. Th electronics too, with ECTS2. Now to get the operations right.

[JimmyP]

Plus, we should be trying to get away from a 'Commuter' network and instead move to a connected network that is used for various trip types by as many people as possible, instead of the current situation where, outside of commuting to the city for work, it's an almost last resort option.

[#Metro]

Access analysis uses 30 minutes as the time budget for a reasonable commute.

Of course commutes can be longer than this, but the proportion of people who want to do that drops.

Marchetti constant https://en.m.wikipedia.org/wiki/Marchetti%27s_constant

This is why, for example, we have express trains to the Gold Coast rather than all stoppers from Brisbane.

You can also use a 60 minute isochrone to model the longer trips.

More information about access analysis here

"Basics: Access, or the Wall Around Your Life"

https://humantransit.org/2021/03/basics-access-or-the-wall-around-your-life.html

[Cazza]

  

It takes me longer than 30 mins to commute by PT the 6 or so kms from Ashgrove to Woolloongabba, so I can see most peoples commutes exceeding this. Good luck getting anyone from their home in Albany Creek, Brighton, Crestmead, Centenary or Forest Lake into their destination in the CBD in 30 mins, especially at peak.

The reality is, SEQ is such a vast, low density and poly-centric region whereby significant numbers of people commute significant distances.

Our heavy rail network is tailored towards this longer distance commuter style, rather than your short haul metro services in European type cities.

It's not unreasonable to expect people here will be commuting much longer than 30 mins, especially with the growing push for regional rapid rail.

Theories are all well and good. But application in the real world is the most important thing to understand and focus on.

[#Metro]

Travel time budgets exist, and are well established in the planning literature. 
This has serious implications for how services should be designed on PT networks, particularly rail.
This is because it affects the value proposition and competitiveness of rail versus the alternatives (car). Which flows on to patronage and market share / mode share.

We also know from comparison of Perth's train lines that the Joondalup and Mandurah lines pull in the most passengers (about 50% of the entire system) for the simple reason that they have very high average train speed (frequency is the same network wide - 15 min, so we can rule that out as a causal factor).

I don't think anyone is disputing that travel time budgets exist. And that in any given population there is a general expectation about what a 'reasonable' average commute time and distance of the home to work.

The average commute, from antiquity to now: half an hour That's Marchetti's Constant, and it informs the growth of our modern cities But we're stuck in traffic, and waiting for the next great leap forward.

https://bigthink.com/strange-maps/marchettis-constant/

Our heavy rail network is tailored towards this longer distance commuter style, rather than your short haul metro services in European type cities.

But it's not tailored. Its offering all-stops LSR and VLSR over very long distances in many cases (e.g. Beenleigh line, Ipswich, & Cleveland). This is a mismatch with the regional low-density urban form we now have.

We are talking about population and group behaviour. We know this from ABS census data.

Around 7.4 million people (or 73% of employed people over the age of 15 years) commuted a distance of less than 20 km to work.

^^ This implies 30-45 minute (average) reasonable commute time.



People do not want to live at a distance that is unreasonable. Which makes sense, because living further means they can't meet their time budget. If trains were faster and more frequent, the viable area and distance they could live in would be wider.

Of workers in capital cities, people in Brisbane (16.7 km) and Melbourne (16.2 km) had the longest average distances to their place of work, while those in Darwin and Adelaide had the shortest (12.8 km and 13.1 km respectively).

Which again, also implies a 30-45 minute "reasonable" travel time budget.

What does this mean?

- Firstly, People have 'reasonable' travel time budgets. PT services that do not fit that will be unattractive to passengers because the value proposition is too poor.

E.g. The Cleveland Line, which is a VLSR service that has an average speed of just 35.5 km/hr and stops everywhere. This is actually slower than driving a car in mixed traffic! Its not a rational choice to take the train in many cases.

- Secondly, it is unreasonable for people to have excessively long commutes. Or worse, to only have one mode (car) that is a reasonable commute option because the alternative is to use LSR or VLSR rail services and blow their travel time budget by 2x or possibly 3x.

- Thirdly, we don't yet have regional rapid rail. We need to move towards that, and making the Ipswich line express all day is the first move. There are other things that can happen:

- Split Cleveland line into two-tier local and all-day express services
- Split Beenleigh line into two-tier local and all-day express services (I would suggest Beenleigh-Kuraby then express to Park Road all day)
- All-Day express services to Kippa-Ring with an infill local service potentially starting at Petrie and then continuing to the CBD
- Infrastructure works to speed up average train times (e.g. Kuraby to Beenleigh Project)

For areas closer in, travel time can be reduced by increasing frequency to 15-minutes all day which reduces platform waiting time.

The current stopping patterns, service offering, and average speed of trains on the current network are completely uncompetitive, and that is before we have even thrown in bus interchange and waiting/service frequency into that. It is time to bring in more ExpressLink services that the Connecting SEQ 2031 plan foreshadowed.

[#Metro]

Jarrett explains it much better: 
https://humantransit.org/2021/03/basics-access-or-the-wall-around-your-life.html

Whoever you are, and wherever you are, there's an area you could get to in an amount of time that's available in your day. That limit defines a wall around your life.  Outside that wall are places you can't work, places you can't shop, schools you can't attend, clubs you can't belong do, people you can't hang out with, and a whole world of things you can't do.

(italics added)

[Gazza]

Do Australians have a higher travel time budget than in the literature, by necessity?

Australia as a whole is bigger, so maybe people are more accustomed to setting aside more of their time to travel.

And cities are bigger as well.

I mean in Europe or Asia you can get a lot of places in 30 minutes.
For example, you can get completely from Amsterdam to the neighbouring city of Utrech in 28 min.

So maybe people don't bother travelling further than 30 minutes because there's no advantage to doing so.... All the shops and potential jobs in businesses you want to visit are going to be within 30 minutes.

Something to think about. I mean even on Perth's great rail system it takes longer than 30 min to get to Rockingham, but Rockingham has a non-zero number of passengers using the station.

........

The point about the Beenleigh line.
This is something the Gold coast faster rail project addresses because passengers will be able to interchange at Loganlea and Altandi, and change from an all-stops to express.

Another point I would like to raise is that many rail lines in Australia are dual track offering an all station service that is a bit slow. Cleveland is an example.

Of course you could benefit those passengers by operating more express services, or by quadruplication etc.

But the cost of those may not be "worth it".

For example, quad track to the Gold coast make sense because you have over half a million people on the Gold Coast alone.

But quad tracks or express services full time to Cleveland might not be worth it because it would be such an expensive and disruptive project to build a quad on what is ultimately a shitty indirect route anyway with too many LXs and Cleveland is not a particularly important hub.

Do you look at the value of a faster alternative rail line instead?

[#Metro]

For example, quad track to the Gold coast make sense because you have over half a million people on the Gold Coast alone.

But quad tracks or express services full time to Cleveland might not be worth it because it would be such an expensive and disruptive project to build a quad on what is ultimately a shitty indirect route anyway with too many LXs and Cleveland is not a particularly important hub.

- Well, it depends on what kind of value proposition you want to offer to passengers. Are we aiming for Pay More, Get More, or Pay Less, Get Less?

- Right now, we have a Pay less, Get Less system. For a low fare/low taxes you can wait up to 30 minutes for a slow train, slower than a trip in a car,  to take you to the CBD.

- It probably does not need to be quad all the way. I doubt it would need to be quad past Manly, for example.

- BCR and NPV measures of project viability are sensitive to investment timing. The same project can be viable or unviable depending on how it is staged, or if it can be staged at all.

- The Gold Coast Line pulls in ~5 million or so passengers per year, and Perth's Mandurah line was pulling in about ~20 million passengers per year pre-COVID, which is 4x that of the Gold Coast line even through the population is about 100,000 people and the Gold Coast is 550,000. Any analysis thus needs to be more sophisticated than just comparing raw population figures.

- IMHO the QLD Rail network is not performing anywhere near the performance of what TransPerth in WA is achieving.  

- Even if we apply adjustment factors for the fact that the rail line only serves 1/2 the Gold Coast, TransPerth's Mandurah line performs far, far ahead of the GC line or to my knowledge, any rail line in SEQ.

- We need a new train service model that is actually appropriate for the low-density regional sprawl that SEQ is. Introducing high-speed Express link trains that omit most inner-city stops and run all day is how to do that. Or build more (tolled) tunnelled motorways. Those are the options.

Something to think about. I mean even on Perth's great rail system it takes longer than 30 min to get to Rockingham, but Rockingham has a non-zero number of passengers using the station.

The average, among others, are measures of the central tendency of a statistical distribution, not the edge of that distribution. Obviously, the number of passengers starts to go down quite a bit after the mean/median/mode. Seeing some passengers at Rockingham station which is 34 minutes from Perth Central is not inconsistent with that concept.


Source: https://www.cdc.gov/csels/dsepd/ss1978/lesson2/section8.html

Trains need to have higher average speed in SEQ, particularly on the existing QR network as in-vehicle time and frequency are the main components of the General Travel Cost Equation that can have the most impact.

The other components of the GTCE are:
- Door to Station access time (~ 10 mins)
- Station to door access time (~ 10 mins)
- Actual waiting time at the stop or platform (varies)
- Any sort of transfer waiting time (e.g. feeder bus to train to station)
- Any sort of mid-connection (e.g. changing trains or changing buses)

When you throw all those other components in, that's the time budget gone for most folks.

[HappyTrainGuy]

Only way to have fast commuting times is longer station spacing and less stops. And yet you want more stops for locals to go short distances. Can't have it both ways. Sure you can have it both was but is it worth the money to do so for the return on investment.

Also, not everyone on the Gold Coast wants to go to Brisbane. The big downside of having multiple large cities is that you can have vastly different commuting patterns than what public transport can offer. Gold Coast to Browns Plains/Darra/Wacol industrial areas for example which is why the Logan motorway is so popular. You can apply a similar method right across the Sunshine Coast to northern nsw.

Edit: the QR network was designed around a freight and goods network from over 130 years ago. Urban sprawl has just resumed many of the properties surrounding stations with many houses built at the time for local workers now being owned by city commuting or demolished for multi-storey appartments. Zillmere had sidings for the meat works. Darra had sidings. Strathpine had sidings. Petrie had sidings. There used to be the Petrie, Zillmere to Normanby shunt movements. Nundah used to have sidings. Tenireffe used to have its own branch network for the docks. Nudgee which was is industrial and farming is slowly morphing into residential with housing estates popping up on the smallest blocks of land. As a result there's no secondary infrastructure such as shops. The same with Springfield, Fitzgibbon and many other areas. Only Springfield, South of Beenleigh and Springfield lines have been designed for passenger use.