A lot of the time when one asks the question “how can we improve public transport in Auckland?” the answer involves spending money. Build a busway, build a railway line, increase frequencies by buying more buses, add a new route and so forth. Of course all these things are necessary to improve public transport, but I also think that perhaps an even stronger focus should be on answering the following question: with the same amount of resources as we current have, how could we make things better? ARTA’s focus on shifting towards a hierarchical route structure is certainly an important step in answering this question, as is the Public Transport Management Act – so that we can minimise the unnecessary duplication of services.
What I do I mean by duplication of services? Well here’s an example, of how the 135/136 bus route pretty much exactly duplicates the route taken by the Western Line train service. A map comparing the route of the two is shown below, with the bus route in blue and the train line in red:
It does seem highly bizarre to me that we have two routes that start in the same place, finish in the same place, largely follow the same alignment and yet both probably receive fairly significant public subsidies in order to operate. Clearly, until we have integrated ticketing the train system and the bus system somewhat have to operate independently of each other, but as integrated ticketing is coming, surely we can use some common sense and reduce situations like this. The resources used to run the 135/136 route could instead be put into a good feeder bus service for a good chunk of west Auckland feeding into perhaps Henderson train station.
A recent post at the excellent Humantransit blog gives some good pointers for trying to get the best amount of bang for your buck, or in that particular case how to cut services in the least disruptive way possible. Nobody ever wants to cut services in an overall sense, and hopefully we will manage to avoid having to do that in Auckland. However, working out the ‘best’ way to cut services can give us some really good pointers for where we should be shifting resources from, and where we should be shifting them to. This is quite an interesting section from that post:
Normally, at moments like this, we hear the tired budgeting metaphor about fat and bone. Yes, they’ve trimmed the fat, but if they have to cut more they’ll be cutting into the bone. (For some reason, there’s never any meat between the fat and the bone in this metaphor, nor many brain cells for that matter.)
I much prefer the metaphor of the transit network as a plant, because a good network redesign often involves some pruning, especially if service has to be reduced. All those overlapping but locally loved services are a lot like what happens to a plant that hasn’t been shaped. It grows too many stems too close together. These stems compete for the limited nutrients and light, and thus none of them has the chance to grow into a healthy trunk. So you have to pick a stem to encourage and cut out the others that compete with it. That’s exactly what the San Francisco cuts did.
Any decent look at an “overall map” of bus services in the Auckland region gives us an excellent indication that perhaps Auckland needs a bit of a clever “prune” to end up in a better situation with a less confusing and more effective system. Obviously we don’t want to actually reduce the amount of services provided in an overall sense, but a shift to having fewer different routes (and in particular fewer competing routes), more direct routes and more frequent services would probably lead to a better system overall.
In San Francisco, the most recent service cuts have actually had a number of silver linings, as Humantransit explains (from another post):
Every transit system gradually acquires odd bits of route that really don’t make any objective sense. They may have been added to take care of some noisy complaint, or they may just be obsolete services that have been superseded by something added more recently. These oddities and complexities tend to hang on because if you try to cut them, the people who are used to them will complain, and there’s often just not that much to be gained — at least in the eyes of our political masters — from enduring these complaints.
So a lean budget season, now and then, can be a blessing, because it creates the political will to fix these things.
That’s the theme of the new round of changes in San Francisco. The lines being eliminated all have some obvious built-in inefficiency or weakness. Usually, the problem is that they overlap the market of other lines to too great an extent, so that the transit system is competing with itself. Wherever you can replace two infrequent overlapping lines with one more frequent one, you’ve usually got a winner in total ridership terms.
The changes reduce the number of routes overall, thus making the system simpler. (Whenever you hear an agency brag about how many routes it has, remember: The number of routes is not a measure of the quality or quantity of your service; it’s a measure of it’s complexity.)
Even better, they’re re-investing some of the savings in higher frequencies on the routes that remain, so that overall mobility isn’t degraded much, and may even improve in some cases.
I think there are a huge number of routes in the Auckland region which could benefit from this kind of thinking. I’ve explained in the past how I think the Pt Chevalier buses could be combined with the Herne Bay buses – creating a better service overall while potentially saving money. The same could be possible by combining the 035 route with the 025/027, or the 006 route with the new “12” route, or putting together the 008 and 009 routes into a half-decent cross-town service along the Mt Albert Road QTN corridor, or the combination of the 135/136 routes with the Western Line as shown above. Undoubtedly there are many other cases around Auckland where this would be possible.
It might seem as though what I want to do is to cut a whole pile of services, but that’s not the point at all. The basic premise is that one very good bus route is much better than two or three very bad bus routes. By combining resources into fewer routes we can end up with very good frequencies that are high enough to encourage people to “turn up and go” (generally a bus every 10 minutes), which will probably end up having even higher patronage than the two combined routes you replaced. Of course there are some trade-offs. People may need to walk slightly further to catch the bus, people may need to transfer between services, but I would hope that these “costs” are outweighed by the benefits of a simpler system to understand and use, more straightforward routes (that are therefore faster) and perhaps most importantly, higher frequencies – so therefore greater reliability and less waiting time.
Perhaps the most important point though is that these improvements shouldn’t actually cost a cent, and might in fact end up saving money. A few other rather simple ideas I have to improve (particularly bus-based) public transport are:
- Wider spacing between bus stops. Fewer bus stops means faster trips.
- Get rid of all “Not in Service” buses. The bus is already travelling, using petrol and using up the driver’s time. The marginal cost of having the service operational to pick up passengers is surely negligible.
- Focus bus priority measures at “pinch-points” on the network, to give bus users maximum benefit over those driving their cars, for minimum cost. Far too often it’s at these pinch-points (yes I am thinking of you Balmoral shops intersection) that bus priority disappears – just when it’s most needed.
- Fewer different routes that are actually almost the same. I don’t know why there has to be a 241, 246 and 247 bus along Sandringham Road when their routes are all almost identical.
- Pre-boarding payment at busy inner-city stops.
I’m sure there are other improvements that could be made cheaply.
Processing...
I had to laugh when I saw some of these bus routes have no marked bus stops and you get on and off wherever. This can hardly make for an efficient bus trip.
I completely agree that a lot of these bus routes should be converted into feeder routes for the trains, perhaps by making them cross town with train connections or through circular loops between a couple of different train stations.
Like you say this is all dependent on integrating ticketing, and along with that a different attitude to PT i.e. one in which transfers are not seen as something to avoid. Again this requires <10 min frequencies on most of these route to be the case.
The psychology of transfers needs to be addressed too. Once integrated ticketing is in place it needs to be shown to be easy to transfer.
One idea could be to compare a transfer to the driving maneuver of turning right* at traffic lights to change roads – a bit of a wait sometimes but really not hard.
(* for overseas visitors, NZ drives on the left so a right turn is across traffic, and in NZ/Aus it’s often specifically signalised by red/green arrow signals)
It is certainly not an impossible task to get people to accept transfering. It seems to work fine in Canadian and Australian cities such as Vancouver, Toronto and Perth. You just need to make the process easy and give an advantage of transfering (like you shift to a much faster vehicle such as a train).
I wonder if anyone has ever asked those that use the bus route why they do so when the train is available? The only thing I can think of is that the bus was more reliable but that should be drastically improved this year once the major track works are complete.
I think we should run a competition with a decent monetary prize to see how much efficiency we can get. This was done recently by Netflix in the US (a DVD rental service) They ended up getting some of the worlds best brains working for them to improve their movie recommendations for a prize of US $1mil. Their solutions had to be tested and proved to be correct. Some teams spent 2 years working on this competition and at the end of the day it wasn’t about the money but about winning as each member of the team that won didn’t get that much money for how much work they put into it.
See http://www.netflixprize.com/
If run and advertised properly I think it would also have the advantage of raising the awareness of PT and getting people to use it
I think the bus route has longer operating hours, and obviously for some people it runs closer to where they live.
Interesting idea about netflix. It would be great to see something like that happen with public transport.
I use the 135 bus and it is a very well used service. It is the only way those who live in Glen Eden South/Kaurilands can get to Henderson/New Lynn or Glen Eden.
Is used by lots of school children who need to get to high and intermediate school.
Sure its useless if you want to go into town- That problem could solved by a busway accross to the Whau River connecting Henderson/Glen Eden it to SH16, but as a local service its excellent.
TopCat, that is exactly the point. If it’s useless for going into town then why the heck does the route go all the way into town? If it was a shuttle/feeder bus between Swanson and New Lynn, then you could double its frequencies with the same amount of resources because the route’s length would be halved.
The problem with transfers is that every time you do it, you eat up a bit more travel time. I worked out how long it would take to catch a feeder bus to the bus station (10 miniutes walk to the bus, at least five minutes wait at the stop, 10-15 minutes to the station, likely at least 5 minutes wait at the station) and on many days, driving to town would be quicker. Wider spacing for bus stops will make for faster journeys for the buses – but it also means that people have much longer walks (more time) and when the weather is bad, this is a big disincentive to catch the bus in the first place. Total travel time (door to door) will always be an important factor when deciding on your mode of transport and often seems to be ignored when the transport planners expect that people will used feeder services and transfers.
Sharon – I think that is part of the point of the post. By rearranging the routes we could get a higher frequency on them which would provide a better service and likely mean lest waiting at transfer stations.
Admin – I think the thing that the big thing that would stop this is public perception, it would viewed as some people as a huge waste of tax/ratepayer money. If setup and run correctly however the benefits would easily out way the cost. A 10% improvement in efficiency from optimised routes would easily save more than $1 mil a year. For Netflix it was so successful that they are looking at doing more for other areas of their business.
The problem with the train service in Waitakere is that large parts of it don’t run that close to the main residential areas. The line was originally constructed to serve the Waikaumete Cemetary and the then small towns of New Lynn, Henderson and Swanson.
Yes you could eliminate the buses running from New Lynn-Britomart though I expect you would need to provide extra services for Pt Chev, Waterview etc.
Well that’s where a Northwest busway would come in handy.
Actually TopCat the line was originally built as the main route north out of Auckland, initally to the head of the Kaipara Harbour at Helensville. The small towns of Henderson et al naturally grew around the stops on the line i.e. the rail line is the reason those centres (and later the suburbs that grew around them) were there in the first place.
If you look at the urbanised area out west south of the Northwestern motorway, it basically follows a linear corridor about 2-3km either side of the rail line. With the right arrangement of bus services no property would be more than 6-8 minutes bus trip away from a station on the line.
North of SH16 there is Massey, Te Atatu, Westharbour etc, that is where a northwestern Busway would come in handy.
Sharon, bear in mind that you are basing you calculation on what it takes now under a system that is not designed to support transfers, either in terms of the timetable, the location of stops or the routing of routes. In fact the various bus lines and the rail line were designed to compete with each other rather than integrate (yay liberal free market economics!), so effectively the are designed to prevent transfers, or at the least not make them too easy.
Total travel time is indeed very important, but this isn’t overlooked by those that support a transfer based system. Effectively transfers add a small amount of ‘null’ time switching between services to give a larger saving of time overall. It’s a little counter intuitive but transfering can make end to end trips a lot faster if the system is designed that way. This is based upon two things that a tranfer based network can provide, better frequencies and a wider range of destinations.
With a no transfer system you have to run all you services from point to point. For this to work you need to maximise route coverage on each line to pick up enough people to make the route efficient, the effect of that is your route tends to wind around the suburbs quite a bit before heading on a more direct run into the city. This wastes time for most people on board.
Furthermore for the system to cost a huge amount of money, you either have to restrict the number of routes leaving from any one point, or you have to limit the frequency. Having a bus from each suburb to every destination in the city every five minutes is simply impossible.
So generally in Auckland you only have one regular bus coming through your suburb which makes its way to downtown, and outside of the peaks it probably only runs every half hour at best. This is another source of delay, if you have to wait an average of 15 minutes for the next bus they you are already 15 mins behind. Apart from a few peak commuter routes to the CBD you are never going to get enough patronage on one point to point route to support more than 30 or 60 minute frequencies, so you can never overcome that first time penalty waiting for the bus. However, if you have a transfer based system then everyone travelling from that suburb could use the connector bus as the first step of their trip, not just those headed to the CBD. This means that the local connector bus might be able to support a service every five or ten minutes which means that wait time is only 2.5 to 5 minutes. As well as using this to access anywhere on the rail network or catch a train to the CBD, at that sort of frequency you could reliably use it to pop down to the shops, head into a regional centre like New Lynn or even visit friends and family in the neighbourhood.
It works at the core of the network too. Right now only those people who live within five or ten minutes walk of a station can use the western line, with perhaps a few more that can get a park-n-ride park somewhere. But say most of the stations on the western line have one of these frequent local connectors. Then all of a sudden basically all of West Auckland is within that five or ten minutes of a station, not just the few percent within walking distance. So with a much larger pool of passengers fast rail frequencies become possible, so a train every five or ten minutes becomes the norm. If you feeder bus runs every five minutes and your train runs every five minutes then the time it takes to transfer starts to become insignificant, particularly in regard to the much wider range of destinations you can access and the faster speeds and greater comfort available on the rails.
Topcat – I think that the rail line gives good coverage to most of West Auckland apart from the areas you mention to the north of city and coincidently around the motorway. I haven’t seen any maps or photos of the area before but I think that the motorway would be largely responsible for this growth.
Admin I fully agree on the busway, it would give almost all of West Auckland decent coverage by and would likely be cheaper and easier to put in than any of the proposed RTN’s in the RLTS. Personally I think we should put it in and watch as Westies flock to it. It will make many residents in the rest of the city jealous and cause them to demand similar systems (particularly out East).
One thing I have noticed is that the Sturges rd station park and ride is not often very full, I think that is mainly to do the fact the station is one extra stage from Henderson. I just hope that when integrated ticketing comes we shake up the zones, this would probably make the station a lot more attractive. A good example is I was talking with a neighbour and found they are dropped off in Henderson just because of the extra cost from Sturges. We are both only a five minute walk from the station.
The point, I guess I was making was that there are very few residential properties within 5 mins walk of any the western line stations
The line runs through a light industrial zone along Railside Av, Parrs Park, the Waikumete Cemetary and Glen Eden, New Lynn and Henderson town centres. Only around Fruitvale and Ranui are there residential housing (not particularly high density), one side of Glen Eden station and one corner of the New Lynn station has a housing development.
Swanson is totally undeveloped to the South of its station, likewise Sunnyvale.
Whilst its good that the town centers have a station, it is unlikely many of the users of the Western Line would be walking to the train given the distances they would need to walk.
What percentage of Waitakere residents would live within 5 mins walk of a station? Not very many I suspect. This figure will continue to fall as Hobsonville and north of Westgate get developed in the next 20 years.
It would be interesting to compare this figure to outher raillines such as in Wellington.
Something for the Land Use planners to think about.
I would actually expect the proportion of people within close proximity of Western Line stations to increase during that time. New Lynn is set to have massive redevelopment and intensification, while Waitakere City has rezoned for high-density housing areas around all its stations.
I guess I am lucky to be so close to the line that I can walk but the western line also has a good amount of park and ride stations to cater for the nearby population. There are parking facilities at Glen Eden (unofficial), Sunnyvale, Sturges Rd, Ranui and I think Swanson now has one. New Lynn could easily gain one if a cap was put on more of the trench.
I think the West would also be the easiest place in Auckland to serve by feeder buses as it is geographically constrained by the ranges on one side and the harbour on the other meaning it is not likely to spread out much more than what it already has and most of the urban area is not far from the rail line or the motorway (if the busway is built)
TopCat – thats why transfers are all that more important – better to speed the whole system up and reduce travel times as a whole than to focus on reducing travel times for a select few.
The main thing we have to focus on with transfers and the system as a whole, is to simplify it, the main problem with our system is it isnt the easiest to follow and there are so many routes that go different directions – even if going to the same place, so you can’t just walk out the door and hop on any bus on the street, you have to actually plan your route’s/day in advance or memorise each different route.
I hope an Auckland wide review is done as part of the integrated ticketing roll out, with 100 odd routes, numbered 1 – 99, mostly as feeders…
If we got rid of that Western bus service following the rail line then how would we move people around in January when we shut down the Western line?
I wanted to go to New Lynn by train today. When I checked Maxx I saw it would take me almost an hour by bus. An hour! I could drive there in 20 minutes. And I will…
Way to encourage passengers Maxx
Bus replacements of trains I suppose. Of course eventually these Western Line upgrades will be completed, as will electrificatio, and we won’t have to shut down the line for most of each January.
After 2013 we shouldn’t have a long January shutdown for 30 years…
I can see the Western line and Britomart needing to be closed at times when we build the CBD Loop as it will be a similar process to what is happening in New Lynn. I also imagine that around Mt Eden we will also need to reduce it down to a single track until the work is complete to get enough space in the railway corridor. Other than that the only line closures should be when we add new lines to the network and as only the junctions would be affected, the affect should only be fairly localised to that particular line.
until we do triple tracking 🙂
Until those projects are announced I won’t be holding my breath…