The MAXX website has a few regional maps that look at all the bus routes in Auckland – split into Central, Northern, Eastern, Southern and Western suburbs. They are useful diagrams in one sense – in that you can find out what routes serve your house and perhaps where you need to get to – although the Journey Planner is pretty useful for that task too. In terms of giving people an idea of what our bus system is actually like though – and what options one has when planning a journey – they are actually pretty useless. That’s because they are incredibly complex diagrams that look something like this:
OK, I realise that’s really really small, here’s a zoomed in look at part of the above map:
Looking at these maps you’d get the feeling that Auckland (at least the isthmus area) has an incredibly comprehensive bus network. But, unfortunately, these maps really only tell us half the story. Most critically, they do not distinguish between a route like the 177, which operates at a frequency of one bus every three hours, and the 267, which has very high frequencies all seven days of the week. So really, this kind of map is pretty useless without the accompanying timetables, to really give us an idea about what a route’s frequency is.
Now generally I don’t really give much of a damn about small distinctions between routes such as the 248 and the 249, they’re all Sandringham Road buses to me. Really, unless you did live at the very end of many of Auckland’s bus routes it would make no difference what the route number is. What matters, when I am deciding whether I should bother thinking about taking the bus for a journey other than my daily commute, is how long I’m going to have to wait for the bus and how slow and windy the trip might be. In short, what matters is service frequency. So, what if we did a map that related to service frequency?
Perhaps in an ideal world we could have a map that make distinctions (perhaps by the thickness of the line) between routes that ran at different services frequencies, but for this task I am going to take a simple “in or out” approach. My criteria is that for a route to make it onto the map, it needs to operate at a frequency during the inter-peak weekday period, of a bus every 15 minutes – or better. I suppose that’s a pretty arbitrary cut-off point, but in my experience it is the difference between a pretty decent service and one that’s not particularly great. Generally if there are buses every 15 minutes off-peak then there will be excellent service frequencies during the peak hour; and also means that during evenings and at the weekend there is probably a reasonably OK service.
The map below shows a pretty rough estimate of the services that operate at 15 minute or better frequencies throughout the weekday inter-peak period. The fact that it was rather difficult to ascertain this information from the various timetables is probably just a further sign that we need to improve the structure of our bus routes in Auckland – as I often had to add up different timetables to work out the total number of services along a route – not particularly helpful for someone trying to quickly work out whether they should bother catching the bus or not. But anyway, here’s the map (and yes I know I haven’t continued many of the routes with 15 min frequencies right to the edge of the map):
My first thought, upon completing this map, was “hey wow it’s Auckland’s old tram map!” Which is a rather sad way of saying “our bus system hasn’t come anywhere in 50 years.” All the best bus routes converge on the CBD, although only 12% of the city’s workforce is employed there, many of the best bus routes directly follow the train lines, all the routes are pretty much designed to serve the passenger from start to finish of their trip (ie. long-haul routes rather than feeder routes) and overall there really doesn’t seem to be much of a co-ordinated system going on.
The blue lines on the map below show how I think the system could be improved, by creating more of a network of routes. It would require integrated ticketing, it would ideally require 10 minute frequencies along many of the routes, and it would require clear and easy transfers between different routes where they cross over. But overall, I think it could result in the system becoming far more useful for people other than those who work in the CBD, and it would become far more useful for a variety of different trips than the system we have now. In other words, it would be a step to becoming “urban transit”, rather than “commuter transit”. A lot of the necessary additional routes are cross-town services. At the moment I think there is a strong belief that cross-town services can never be viable at particularly good frequencies. Probably the best cross-town service at the moment is the 007 route, which runs at semi-decent half hour frequencies during the weekday inter-peak, but utterly horrific “almost two-hourly” frequencies on a Sunday. Reducing the over-supply of free parking in areas outside the CBD is also a very important step in making cross-town public transport viable, but I do think at the moment a large part of the problem is the terrible service provided.
With integrated ticketing, higher frequencies on this “core” part of the network and more bus lanes to speed up travel for buses along this part of the network particularly, I think that public transport really could become a viable option for many more people than simply those who work in the CBD. With only 12% of Auckland’s population working in the CBD, it is essential that we work on stuff like this to reduce the auto-dependency of the other 88% of our working population.
Processing...
I had been thinking about something similar after seeing these area guides, particularly how can you map both the route and the frequency on one page to get the true view of the network.
I though about varying the line thickness, but given the great variation in service levels you would have some lines paper thin and others covering half the ithsmus! The solution I thought was to map each individual bus serivce along its route on a graphics program with 3D capabilities (polygons in Google Earth perhaps) and then give it a height based on the hourly frequency (say 1mm for every service per hour). Where multiple bus routes share the same road, their individual lines are stacked upon each other. So you could look at Great North Rd and the line might be 30mm tall indicating 30 buses an hour running up and down the street, while over on the 007 the line might only be .3mm tall.
It wouldn’t map the temporal dimension, so you might need to do say three versions, one for the peaks, one interpeak, and one evening.
To go one step further than that would be to account for carrying capacity i.e. some way to show that a half hourly train and a half hourly shuttle bus have vastly different service levels.
Crikey that sounds complex Nick! I think the whole idea of this is to give people a concept of where the “decent bus services” are.
Great analysis Admin, there are literally many times the gains to be made by improving bus services than the rail improvements, many can be done cheaply and easily right now…
Simplifying Nick’s idea a bit…
Solid lines for 30 min or better inter-peak and weekend frequency (whichever is lower), dotted lines for anything less frequent.
Thick lines for 15 min or better peak frequency, thin lines for anything less frequent.
So a half-hourly at peak bus would be thin and dotted, Dominion Rd thick and solid.
The Beachhaven Loop would be thick – but dotted (40min weekend frequency is not good enough)
A big problem is that once you get out into the post World War II suburbs we abandoned the “super grid” for arterial routes, which means that it’s nigh on impossible to avoid looping indirect bus routes.
I believe there needs to be a number of district hubs where numerous bus routes converge, so people can change onto other bus routes. In many places these should be railway stations where the railway station is appropriately close to a town centre. On the central map I would have a quick guess that New Lynn, Onehunga, Otahuhu and Panmure could be major hubs, with Avondale, Mt Roskill, Glen Innes being smaller hubs.
Services between these hubs would then run at 10min frequencies, and then local services could run connecting the low-density suburbs to the hubs, although these would have to be at lower frequencies. Although increasing the number of destinations that can be served increases the viability of these suburban services, and vice-versa.
I agree Luke. The obvious points are along the 007 route at the moment, which I think could be a really useful route in linking together a great number of trips – if it had better service frequencies and was more reliable.
Proper transfer points at St Lukes, Balmoral, Epsom, Greenlane, Remuera & St Heliers would really tie the whole system together.
Wouldn’t be nice to be able to cycle to a bus stop on a high frequency route, park your bike (at one of the racks provided next to the bus stop) and then catch the bus to wherever you want to go…
I was meaning for the purposes of analysing the network to work out where improvements should lie, not as a map for the public!
My way of organizing the system would be for each area to reciveve a two digit number and a colour on the map (for example LINK would be 01 and light green, Western bays 02 and dark green). Each variation would be lettered and the letter would be marked at the terminus of the route. Not sure how to deal with cross towns that service a number of areas
I think we need fewer different routes and to have them run at higher frequencies. There will be a tradeoff in that people might have to walk a bit further to their bus stop – but I think that would be more than made up for by the reduced waiting time (through better frequencies).
I think the main roads where there is a parallel railway route should be the first ones that need rationalising. Should one service really be running all the way from say Red Hill, a suburban part of Papakura, all the way into Britomart. With such a long route I can’t imagine it keeps time very well to midway stops such as Manukau City. These would mean a long and frustrating wait for those at midway points. Would be much better to have two seperate services with a split at Manukau City. Both these services should be run an 10min intervals, which would also mean the bus could act as a rail feeder as well as being useful for short and medium distance trips.
Apart from Great South Road, I would see Great North and New North Roads as being candidates for similar treatment, with a split at New Lynn.
Luke, that is why integrated ticketing is so essential. I imagine there are plenty of people who live down south or out west who would prefer to catch the train, but aren’t within walking distance of a train so basically have to catch a really long bus trip.
It’s also an incredibly wasteful use of resources having all these duplicated routes.