How do we improve the cost recovery of Citytrain (or Translink)?

[rtt_rules]

Guys,
Recently I read the ministers defence on why Translink fares are rising so much year on year (which I think has by-partisan support). Aim is to reduce the subsidy from 70 to 75%. A few years back before Translink I thourght the BCC stated Citytrain cost the taxpayers 65% but the BCC buses was 60%, so costs must have risen a far bit. But suprised that the proposed fare rises only get back another 5%.

This old article from CM (mid 2010) tells a worse story and using their numbers in mid 2010 it was close to 80% subsidy. I cannot find any comparisons, but I would have put 70-80% subsidy as being very high. Also makes expansion a disincentive to govt as adding more stations or more trains (ie 15min timetable) just adds more cost to the annual general revenue.

So my question to all is, How do you fill the gap Increase revenue/decrease costs? The end result is a system that is more affordable to the bean counters in the govt and hence more attractive to expand.

Regards
Shane

[Gazza]

Retrain guards as drivers and cut admin staff.

[SurfRail]

Run more services in the off-peak to attract more ridership.

[Mr X]

Cut out bus stops to make welfare routes faster
Remove useless air parcel routes
Reduce numbers of duplicated school bus routes

[somebody]

Run more services in the off-peak to attract more ridership.

This is the main game IMO.  DOO is secondary.

[#Metro]

The proportion of duplication (ie P88, 461, 416, 476 etc) needs to go.
There are some welfare routes that should be cut because they are duplicated (198)

People need to LET GO, grieve the loss of deadwood, and move on.

As for CityTrain, the main issue is that most of the patronage in the off peak is air-parcel services. Two way frequency is terrible and just doesn't make sense to use the services. And this is true between major destinations as well - i.e between Ipswich and Brisbane CBD.

[ozbob]

The bus network needs to be remodelled on core trunk routes, bus and rail, feed by frequent feeders over a wider span of hours.

Off peak, I travel a lot and there is good loadings generally, but can be improved by increased frequency, with better bus connections. And further off peak discounts to lock it in.  Maximising public transport use around the clock is the way that the fare box will be improved relatively.

[morb]

Run more services in the off-peak to attract more ridership.

You're telling me that the cost to establish the extra driver shifts, more frequent maintenance intervals, etc would be much more than recouped by the extra patronage?

[morb]

The proportion of duplication (ie P88, 461, 416, 476 etc) needs to go.

In the context of value for money, how are these routes detrimental?

P88 for example, takes pressure off Cultural Centre.  And we all know the mega$ solutions being touted to take pressure off Cultural Centre.

The whole Cultural Centre thing is quite odd, actually.  For a while there were ticket sellers in peak hours and last I knew they'd been taken off.  Why is that?  Surely it would have taken the pressure off drivers having to stay longer at CC to sell tickets on board?

The other problem that would seem more intractable are the 2 traffic lights when passing under the railway.  Things get quite jammed up there.

[#Metro]

When you double the frequency, you double the cost. This is true, because frequency (labour) is expensive. (The exception is automated systems).

On the other hand, when you double the frequency and expand the service hours, the patronage doubles too. We know this from the BUZ data that minimum increases
over a number of years sees at least a doubling in patronage.

Furthermore, with a better quality network, there is also more moral weight to charge for it too.
I'm happy to pay 15% more to use the busway and the BUZ because it saves me time.

The final reason is Core Frequency. Once there is decent capacity on the network, you can feed buses to it rather than run them all the way to the CBD.

For example, the metroad 5 corridor. That could get a BUZ 359 and 594 on it and perhaps retain a 350 Everton Park service. Everything else could be terminated at Ennoggera and people loaded on to rail services.

Rail systems have a lot of infrastructure and fixed costs to maintain, whether they carry 1 million passengers or 1 passenger.

[#Metro]

In the context of value for money, how are these routes detrimental?

P88 for example, takes pressure off Cultural Centre.  And we all know the mega$ solutions being touted to take pressure off Cultural Centre.

P88 does little to take pressure off Cultural Centre. One bus holds 65-85 people, so in one hour, running the service every 15 minutes, only 340 people are diverted away from there
which is 3.7% of what Cultural Centre Capacity does at the moment. This bus route costs millions and millions of dollars to run, and for what? To save people a 30 second walk.

In any case it is not enough to prevent Cultural Centre from going overcapacity, which is exactly what has happened.
It would be better perhaps IMHO if 160 was merged with P88 and was sent over the Captain Cook Bridge instead and extended to Roma Street, and the western leg of
P88 dissolved to form part of BUZ 400.

The whole Cultural Centre thing is quite odd, actually.  For a while there were ticket sellers in peak hours and last I knew they'd been taken off.  Why is that?  Surely it would have taken the pressure off drivers having to stay longer at CC to sell tickets on board?

I think there were replaced by the ticket machine and security to check tickets.

[BrizCommuter]

Cut the fat in QR. A very bloated organisation (as with most government organisations), and that's before we even get to the guards!

[somebody]

Run more services in the off-peak to attract more ridership.

You're telling me that the cost to establish the extra driver shifts, more frequent maintenance intervals, etc would be much more than recouped by the extra patronage?

Technically, it doesn't need to be more than the extra revenue.  If the extra revenue is more than 25% of the costs or so there is an improvement in the farebox ratio.

Personally, I believe that the extra revenue would most likely pay for the marginal costs, but I could be wrong.

[morb]

On the other hand, when you double the frequency and expand the service hours, the patronage doubles too. We know this from the BUZ data that minimum increases
over a number of years sees at least a doubling in patronage.

Just wondering where those figures are published?  I'd like to check for confounding factors, such as the ramp up in petrol prices in 2007, and whether non-BUZ routes have had the life sucked out of them by the BUZ routes.

Furthermore, with a better quality network, there is also more moral weight to charge for it too.
I'm happy to pay 15% more to use the busway and the BUZ because it saves me time.

Problem is you pay more whether you're on a BUZ or not.

The final reason is Core Frequency. Once there is decent capacity on the network, you can feed buses to it rather than run them all the way to the CBD.

One thing I've noticed about the discussions here are a lack of acknowledgement of the "1 trip" factor.  I remember Maureen Hayes saying, essentially, people love single trip (she also mentioned they hate more than 1 hour of commuting time). 

I understand the lure of your system.  Given my commuting style though, it seems like going backwards.

First there is the problem of waiting for the bus.  10 minute frequency still means I am twiddling my thumbs for 5 minutes on average (and even that is optimistic given the "3 buses arrive at once" effect).  If I read a book or smartphone I am still worried about whether I neglect to hail the bus in time.  I'd hate to be at Cultural Centre on an afternoon peak, you could never relax.

One solution to this is to have real time bus info - I know Translink have alluded to this in their annual reports but there seems to be little transparency to how the requirements are being gathered.

It would be truly great to have a smartphone app that gives you a 30-second (or whenever) heads up for when your bus is actually arriving!  Gives me a moment to put the smartphone away, check for my go card, hail the bus.

My second problem is the interchange amenity.  For me, RBWH is an acceptable enough amenity.  Lutwyche is not.  Too loud, water splashes on you when it's raining, yuck.  To run more of a grid or feeder system would require a lot more attention to the interchange points.  Maybe this is still worth it, I don't know.

[#Metro]

What fat? Needs to be more specific.

A quick personal index of mine to see how efficient an organisation is, is to simply look at the staff : customer ratio.

Queensland Rail is a bit hard to compare in the sense that it operates a network much larger than just the city passenger rail system.

But anyway, QR has 7000 staff. Let's say half of them are on the passenger network. So 3500

There are around 85 000 passengers who use QR every week day (derived from knowing that 170 000 trips on rail are done on weekdays).

So that's 24 passengers per staff member, under this working assumption.

Brisbane Transport has around 2000 staff members. They carry 70 million trips per year so div 52 weeks div 5.5 days a week div two journeys is 122 377 passengers per day.

So that's about 61 passengers per staff member.


Now does this mean that everyone should be on buses? Not necessarily. One interpretation could be that the ratio needs to shift - passengers need to go up and staff down (or at least be held constant - i.e. convert guards to drivers which would be labour neutral).

[somebody]

On the other hand, when you double the frequency and expand the service hours, the patronage doubles too. We know this from the BUZ data that minimum increases
over a number of years sees at least a doubling in patronage.

Just wondering where those figures are published?  I'd like to check for confounding factors, such as the ramp up in petrol prices in 2007, and whether non-BUZ routes have had the life sucked out of them by the BUZ routes.

This link: http://www.google.com/url?sa=t&rct=j&q=thredbo%20buz%20warren&source=web&cd=2&ved=0CDwQFjAB&url=http%3A%2F%2Fses.library.usyd.edu.au%2Fbitstream%2F2123%2F6058%2F1%2Fthredbo10-themeA-Warren.pdf&ei=q4kbT6KyEomciQefiOmgCw&usg=AFQjCNHux8eVG0D8SieGJyoL-n_jM-wU9w&cad=rja

Has something on the 444 vs other routes on the corridor.  It shows that the 444 increased the corridor quite a bit, and it seems that the other routes improved thanks to the 444 being there.  Of course, it is only up to 2007, and in 2008 they thought it would be fun to break the corridor (444 into KGSBS).

[#Metro]

Just wondering where those figures are published?  I'd like to check for confounding factors, such as the ramp up in petrol prices in 2007, and whether non-BUZ routes have had the life sucked out of them by the BUZ routes.

The main cost of PT is paying the driver. If you double the services, you double the time worked by drivers thus you double the cost. Other factors will result in variation but overall will be outweighed by labour costs. See Human Transit Book by Jarrett Walker.

Problem is you pay more whether you're on a BUZ or not.

This is true, but we are talking about a network of services rather than individual lines. Rail users will actually see a degradation in value for money (VFM) because the price has gone up but the frequency hasn't. Bus users on the other hand have seen an increase in VFM because things like BUZ 412 / BUZ 180 / P88 / BUZ 222 / Route 29 / BUZ 100 and so forth have been made possible by that fare increase. Non-BUZ users benefit because they can change to a BUZ and not wait so much. Also, once the core of frequent services are in, these lower frequency routes can be terminated at suburban shopping centres for frequency upgrades at little or no cost.

One thing I've noticed about the discussions here are a lack of acknowledgement of the "1 trip" factor.  I remember Maureen Hayes saying, essentially, people love single trip (she also mentioned they hate more than 1 hour of commuting time).  

I understand the lure of your system.  Given my commuting style though, it seems like going backwards.

1 trip journeys are appropriate for smaller cities and towns. The problem with the 1 trip is that you have buses from every suburb going to the CBD all duplicating eachother and carrying large amounts of air to the CBD. That costs a lot of money. It also puts phenomal pressure on the core - CBD streets and spaces, which require costly infrastructure upgrades. The reverse problem is that buses fan out into the suburbs which dramatically dilutes frequency to half hourly or even hourly, which is very unattractive. Connections allow services at more attractive frequency to more destinations and also faster trips. It also means a simpler and more legible network. http://www.humantransit.org/2009/04/why-transferring-is-good-for-you-and-good-for-your-city.html

Transfer penalties can be reduced to almost nothing, which is what I observe at Cultural Centre when I transfer- the bus I want is usually already there.

My second problem is the interchange amenity.  For me, RBWH is an acceptable enough amenity.  Lutwyche is not.  Too loud, water splashes on you when it's raining, yuck.  To run more of a grid or feeder system would require a lot more attention to the interchange points.  Maybe this is still worth it, I don't know.

Suburban interchange needs work, but it is possible. CityGlider like bus stops could be put out with a large sticker saying "Interchange stop".

[morb]

Great topic, rtt_rules.

P88 does little to take pressure off Cultural Centre. One bus holds 65-85 people, so in one hour, running the service every 15 minutes, only 340 people are diverted away from there
which is 3.7% of what Cultural Centre Capacity does at the moment. This bus route costs millions and millions of dollars to run, and for what? To save people a 30 second walk.

And save the hassle of looking out for your 2nd bus, as per my previous post here.

Anyway, we're talking about peak, where P88 runs closer to every 10 minutes.

In any case it is not enough to prevent Cultural Centre from going overcapacity, which is exactly what has happened.

Since I'm not a regular CC user, did the overcapacity happen pretty much suddenly one year?  (Queuing theory predicts this, by the way.)

It would be better perhaps IMHO if 160 was merged with P88 and was sent over the Captain Cook Bridge instead and extended to Roma Street, and the western leg of
P88 dissolved to form part of BUZ 400.

Which way would 400 go?  Via Roma St?  Might make sense if the preferred interchange was KGS.

The whole Cultural Centre thing is quite odd, actually.  For a while there were ticket sellers in peak hours and last I knew they'd been taken off.  Why is that?  Surely it would have taken the pressure off drivers having to stay longer at CC to sell tickets on board?

I think there were replaced by the ticket machine and security to check tickets.

Does that mean no ticket sales are required on-bus?

[#Metro]

And save the hassle of looking out for your 2nd bus, as per my previous post here.

IMHO P88 is not really a significant mobility improvement.

* It's relief of Cultural Centre is very tiny (5 %, assuming 10 min frequencies in peak and 100% full buses)

* Any capacity freed up at CC will immediately be eaten up by other services - it is not enough to prevent overcapacity anyway,
so while "taking pressure off CC" is technically true, it is not significant. BUZification of 180, 100 and 222 have already wiped out any
capacity freed up by P88 anyway.

* Coronation Drive is already a high frequency corridor with buses every 30 seconds - 1 minute, and less than every 5 minutes at other times.
P88 thus does not reduce waiting time significantly along this section and also duplicates everything else- placing frequency exactly where it is
*NOT* required.

* The cost to provide this service runs into the millions and millions of dollars per year, on a scale of priorities, money that would be better
spent on extending mobility to suburbs that currently do not have access to a BUZ (Bulimba - 230 or Centenary Suburbs - 400 or Albany Creek - 359)
rather than save someone 30 seconds.

Which way would 400 go?  Via Roma St?  Might make sense if the preferred interchange was KGS.

Ideally via Roma Street when there are decent connection opportunities with both rail and other busway services.

[SurfRail]

Run more services in the off-peak to attract more ridership.

You're telling me that the cost to establish the extra driver shifts, more frequent maintenance intervals, etc would be much more than recouped by the extra patronage?

Yes, that's what I'm saying.  

Doubling frequency = doubling the cost.  However, patronage growth on BUZ services has been shown quite conclusively to follow a higher trend than double patronage.  You don't just get double the passengers - you get something like 3-4 times the passengers, all paying the same fare as the original bunch.  That makes your service much more efficient.

The people in charge of the money either do not understand this, or are ignoring it.

[#Metro]

Human Transit costs --->  http://www.humantransit.org/02box.html

[#Metro]

The people in charge of the money either do not understand this, or are ignoring it.

They seem to know for buses but not trains!

Also, a CFN allows air to be expelled from other services when they are feederised.
This also represents a gain in efficiency, provided the money saved is folded back into new services to increase the frequency and cut waiting time.

[SurfRail]

They seem to know for buses but not trains!

One thing you can give credit to BCC for doing is implementing BUZ as a basic standard.  The state may pay for the buses and the services, but they wouldn't have done something as impressive initially (would no doubt have been 15 minutes 6am-9pm only or something like that).

[ozbob]

Another variable, passenger/kilometres   rail > bus,  this also is a factor.

As posted earlier, fill up the trains, costs the same to run a train for 10 pax as it does for 500 ...  in a general sense.

Fill up the trains by increasing the frequency and redirecting air to stations rather than CBD.  Darra to Northgate has had 45% increase already I understand ...

I was at Darra late last week, parked out ++++, worse than PR  (prior Richlands ...)

[somebody]

Doubling frequency = doubling the cost.  

You're ignoring the fixed costs, which for a rail system are very significant.  Less significant for buses.

However, patronage growth on BUZ services has been shown quite conclusively to follow a higher trend than double patronage. 

The people in charge of the money either do not understand this, or are ignoring it.

Not sure that is true.  I've done some calcs WRT the 444 vs 440 before it, and for every doubling in frequency, it got about an 80% increase in patronage up to 2007.

[somebody]

Since I'm not a regular CC user, did the overcapacity happen pretty much suddenly one year?  (Queuing theory predicts this, by the way.)

Eastern Busway opening happening in August had an effect on peak hour.  Previously AM Old Cleveland Rd routes used to access the Captain Cook Bridge via Stanley St, now they need to use the busway, which increases queuing at the Allen St "loop".  Not directly related to the CC, but has an effect on the possibility of shifting routes away from the CC.

[morb]

However, patronage growth on BUZ services has been shown quite conclusively to follow a higher trend than double patronage. 

The people in charge of the money either do not understand this, or are ignoring it.

Not sure that is true.  I've done some calcs WRT the 444 vs 440 before it, and for every doubling in frequency, it got about an 80% increase in patronage up to 2007.

Hmm, if I'm reading you right, those figures are going the wrong way for improving the cost recovery (I would have expected 100+%).

I suppose we don't have figures for passenger number trends in peak hour (where you would need to add to the fleet) and off peak?

Another confounding factor is, if I recall, CNG buses need to be refuelled between peaks.  That may complicate the ramping up of off peak services.

In the bigger scheme of things, the way the employment is laid out and shift times etc., helps.  Then you have more opportunities to carry pax on counter peak services, and/or spread out rosters during the day.

[HappyTrainGuy]

What fat? Needs to be more specific.

A quick personal index of mine to see how efficient an organisation is, is to simply look at the staff : customer ratio.

Queensland Rail is a bit hard to compare in the sense that it operates a network much larger than just the city passenger rail system.

But anyway, QR has 7000 staff. Let's say half of them are on the passenger network. So 3500

There are around 85 000 passengers who use QR every week day (derived from knowing that 170 000 trips on rail are done on weekdays).

So that's 24 passengers per staff member, under this working assumption.

Brisbane Transport has around 2000 staff members. They carry 70 million trips per year so div 52 weeks div 5.5 days a week div two journeys is 122 377 passengers per day.

So that's about 61 passengers per staff member.


Now does this mean that everyone should be on buses? Not necessarily. One interpretation could be that the ratio needs to shift - passengers need to go up and staff down (or at least be held constant - i.e. convert guards to drivers which would be labour neutral).

You can dive into so many internal organisations that are operated by QR but you would have to be more specific if you want to cut fat like you said. Prior to the split there were 5 dominate structures being Freight, Coal, Network, Passenger and Services which conducted a heck of a lot of jobs all over Australia for other governments and other rail networks and operators but has since lost the Freight, Coal and some of the Services operations. The passenger side of things (last I heard) was made up of about 3000 people which cover promotion/marketing, safety (RailSmart etc), the heritage rail side of things, CityTrain, TravelTrain and TouristTrain all across Queensland. Network Operations has about 800 people that keep trains moving across most of Queensland. QR service has about 3000 people which is made up of maintainence crews for rollingstock, landscaping crews, cleaners (Vandalism/rubbish etc), engineers, electricians, welders and so on and so on which cover and look after all 7000km of track across Queensland still looked after by QR.

[somebody]

Not sure that is true.  I've done some calcs WRT the 444 vs 440 before it, and for every doubling in frequency, it got about an 80% increase in patronage up to 2007.

Hmm, if I'm reading you right, those figures are going the wrong way for improving the cost recovery (I would have expected 100+%).

Not if more than 10% of the initial costs are fixed costs, like bus depreciation.

[SurfRail]

You're ignoring the fixed costs, which for a rail system are very significant.  Less significant for buses.

Yes I am, because they are irrelevant for the purpose of discussing improvements to frequency (unless you actually need more infrastructure to support it, which is generally not the case for us).  As you say, the line is there whether used at 30 minute headways or 5 minute headways, so the fixed costs can be ignored.

Not sure that is true.  I've done some calcs WRT the 444 vs 440 before it, and for every doubling in frequency, it got about an 80% increase in patronage up to 2007.

Sunday patronage has gone up something like 200% on some BUZ routes even though the baseline frequency was only doubled.

The Perth figures also show what happens when you put in frequency and reliability (ie Mandurah line).  Massive increases (now at 18 million a year).  Previous patronage on the freeway buses was nowhere near that.

Even at the same cost recovery figures, it's still a much better use of the funds than investing in things like new M1 interchanges.

[david]

I was at Darra late last week, parked out ++++, worse than PR  (prior Richlands ...)

Hence why extending the 467 into Jindalee and Mt Ommaney is a BAD BAD idea. It will just add to the number of air parcels running around. Diversion of the 467 to Darra is required, to provide and attractive service and to (hopefully) relieve the parking situation at Darra.

On a side note, any reports on how Oxley is looking these days? I went past a couple of weeks ago and there was some street parking available, albeit quite a walk from the station itself.

[Cam]

Cut the fat in QR. A very bloated organisation (as with most government organisations), and that's before we even get to the guards!

Spot on! However, that won't be popular amongst some members of RBOT who are employed by QR because they may lose their jobs.

Unfortunately, the current Queensland government took the option of increasing fares about five times CPI for 5 years in a row instead.

[somebody]

Sunday patronage has gone up something like 200% on some BUZ routes even though the baseline frequency was only doubled.

Old 440 was hourly, and had shorter operating hours.  So that one has more than quadrupled, for a 266% increase in patronage.

[#Metro]

Spot on! However, that won't be popular amongst some members of RBOT who are employed by QR because they may lose their jobs

If you value waste and inefficiency, that obviously has the consequence of needing to be paid for somehow!

That said, I haven't seen or heard any specific and details about waste and inefficiency in QR yet. So I'm not entirely convinced. The exception would be DOO.

[somebody]

- Frequency, I agree more trains will encourage more users. Would be good to see some sort of comparison with Perth. All I can find is in WIKI TP is 360,000/day and Citytrain is 60mpa, which makes Citytrain half has popular (unlikely).

Perhaps that's across modes.  But wiki tells me transperth trains get 54.75mil train boardings in 08/09, and bus is 73.55mil in 08/09.

[BrizCommuter]

Spot on! However, that won't be popular amongst some members of RBOT who are employed by QR because they may lose their jobs

If you value waste and inefficiency, that obviously has the consequence of needing to be paid for somehow!

That said, I haven't seen or heard any specific and details about waste and inefficiency in QR yet. So I'm not entirely convinced. The exception would be DOO.

For starters QR can cut the pointless "customer service representatives" who travel around the network opening doors for people and usually look completely bored.

Given the number of voluntary redundancies offered by QR National post privatisation, BrizCommuter would expect that QR Passenger is equally as bloated.

[dwb]

Introduce a competitor?

[#Metro]

I think guards should go- perhaps they could become drivers and services boosted.

Station staff during the day as well - busway stations don't have them so I don't see why other places need them. I know this is opposite to views on this board, but that's what I think. If there are safety issues, then something could be done after 5pm for those stations.

I am reminded of a story in Human Transit, and I think it is pertinent to the QR situation.

p79

"Caltrain, however has a long history of operation with labour-intensive and therefore infrequent commuter rail. The midday service is hourly at this writing, which is useless for the spontaneous trips that true rapid transit would allow." Jarrett Walker, Human Transit (2012).

Do you want frequency or do you want 'the human touch'.

I know which one I would prefer!

[dwb]

The higher the frequency the less needed the station staff, however, having been on various systems overseas, staff can make all the difference in the world

[#Metro]

The higher the frequency the less needed the station staff, however, having been on various systems overseas, staff can make all the difference in the world

The less need, but if you have a lot of pax who pay you can afford to have lots of staff.