The patronage results for May are out and again the numbers are increasing – although not quite to the same level as recent months. This is in part due to there being one less business day in the month compared to May last year. Here are the results.
Once again the rail network is leading the growth with an over 12% increase in patronage compared to May last year although AT say if that is normalised to account for the reduced business day it increase is actually 17%. That’s fairly impressive considering just how poor the performance of services has been – more on that soon. The primary driver for patronage growth continues to appear to be on weekdays with AT saying there are now around ~48,000 trips on a weekday on the rail network which is up from ~41,000 a day in May last year.
The other normalised results are:
- Total – 6.2%
- Northern Express – 13.3%
- Other Bus – 3.3%
- Ferry – 8.3%
With the continued strong growth in the busway it once again highlights that focusing on rapid transit services is the right approach. Combined rail and the NEX services know make up 21% of all patronage across the network and that figure is growing fast. While many areas of the PT system are obviously in need of improvement, the strong growth in the RTN is a message I really do hope is getting to the Minister as RTN’s are the PT equivalent of motorways and really the kind of infrastructure the government (and of course AT) should be investing in.
With the Other Bus patronage a bit lower than the other modes, I wonder if that was impacted by the decision by AT to start charging for the City Link Bus (previously free with a HOP card).
Coming back to the issue of trains, as mentioned growth has been very strong despite an appalling service standard lately. Out of just over 12,000 trains that were meant to run in May, 650 – (or just over 5%) of them were cancelled – or at least didn’t reach their final destination for some reason. On the Western Line around 10% of all services didn’t reach their destination although I suspect many of these were cases of trains terminated at Swanson. Of those that did run around 20% ended up late. That’s a slight improvement on the month before but still dismal. I guess it proves that passengers will put up with a lot of disruption but likely only for so long.
AT say that five services across the rail network exceeded their planned standing/sitting ratio. This has commonly been reported however interesting one Eastern line service is mentioned which highlights just how very popular there the electrics are in driving up patronage.
Bus performance isn’t quite as bad – although it too could always be better. This sis shown in the table below
One good thing AT has recently done in is start publishing patronage data in on their website in .xlsx or .csv format without people having to trawl through years of documents. I’m told this is just the first step and that more data other than patronage will be coming over time. This is nice to see.
As well as patronage, HOP usage continues to increase and AT say that 72.4% of all trips were made with HOP which is up from 67.8% in April. I’m guessing the fare changes helped with that boost.
Lastly the data for May available yet however here is the results from Wellington up to April. Bus patronage continues to bob around the 24 million trips per year mark however rail patronage is numbers are increasing with April seeing annual growth of 5.5%.
Processing...
Good that rail growth is continuing to hold up despite the utterly appalling service over the past few months. I am guessing that many passengers are losing patience though and I wouldn’t be surprised if ridership growth slowed down a bit before hopefully electrics and a new rail operator fix things.
Maybe, I think most people, even when seriously inconvenienced, understand the big changes the service is going through and how hard that must be, and as they experience the new trains can see the point. But I suspect this patience certainly has a limit. And that limit might just be the full roll out of the EMUs. So the pressure is on.
Mark Lambert was quoted in the Herald as saying the new trains will take about a year to ‘bed-in’. He may be right but I don’t think he should expect the public to be happy for it to take that long.
At least a year…..
International experience suggests considerable time does need to be allowed for to iron out the bugs. At the end of that process, assuming the design, engineering and manufacture/installation is fundamentally sound, the distance between failures should start to really push out. The best examples for EMUs achieve 100,000km between faults or more. The challenge for Auckland is that the trains can’t be looked at in isolation. Reliability is absolutely dependent on the overhead wires, the track, the signalling, the trains and how well they all work together. As it is a brand new network, there isn’t an easy existing reference benchmark for comparison.
Yes I assume he is talking about the whole interconnected system. And yes it’s not technically unreasonable, or unlikely, but it is long for the customer.
We can only hope that they are all very very focussed on this.
Well, EMU’s have been online for a year or more, so should be well and truly ‘bedded’ in by now. Obviously this is not the case.
I’m not so sure Patrick. You have to remember, these latest inconveniences are just the latest in years and years of inconveniences. This is not a new thing, rather a particularly bad version of ‘normal’.
I don’t buy the bathtub argument as we have never actually experienced a sustained period of reliability on the trains.
Are the bus punctuality details now using the HOP card data? First time I’ve seen non 99%+ rubbish, but it still seems a lot more reliable that how I always experience it on the ground, perhaps I’m unlucky or just that it’s the times that buses never show up that stick in the memory.
They’ve been using GPS data for a while now however punctuality is still based on whether they depart on time not when they turn up.
“Reliability and punctuality statistics for bus services are based off the number of sighted scheduled bus journeys during the month.”
I assume ‘sighted’ refers to the fact that their GPS system is horribly unreliable and doesn’t register all trips.
They only count the buses (a) showing up to start a scheduled run as expected and (b) that bus starting its run within a set time frame.
Heres the quote from the AT blurb on bus reliability stats:
“For May 2015, 96.84% of total scheduled service trips were operated, leaving their origin stop no more than one minute early or ten minutes late (reliability
measure). Service punctuality for May 2015 was 92.11%, measured by the percentage of total scheduled services leaving their origin stop no more than one minute
early or five minutes late (punctuality measure). Reliability and punctuality statistics for bus services are based off the number of sighted scheduled bus journeys
during the month. Statistics from 1 July 2014 are AT-reported using GPS-tracking data comparison to scheduled times, and no longer self-reported by bus operators.”
This is quite different from trains because they record start times and arrival times as per this:
“Service delivery (or reliability) is the proportion of trains that are not cancelled in full or part and arrive at their final destination. Punctuality is the proportion of trains
that were not cancelled in full or part and that arrived at their final destination within five minutes of the scheduled time regardless of whether the train departed its
origin on time.
For May-2015 service delivery was 94.6% and punctuality was 80.3% compared to the 12 month average of 96.3% and 84.7% respectively. ”
So you’re not alone with your experience, measuring only that the bus left and how close to the timetable it was when the bus left but not when it arrived is pretty crap, as the arrival time reliability is just as, if not more, important than.the departure stats.
The buses will be on time through the day and in the evening when few people are on them and the roads are less busy. I’d like to see a measure of 7am-9am and 4pm to 6pm reliability. Those are the times if day that really matter to a large proportion of users I suspect.
The December timetable change really exposed the many choke points in the rail network, namely Britomart, Newmarket and Westfield. The extra services also mean shorter turn around times at Britomart (6 minutes at peak hours) and Papakura (8 minutes) . Both factors means almost no fat in the system when a train runs late and causes a domino effect for subsequent services.
And I believe the Manukau services have an unrealisticly quick trip time based on the best possible scenario. I think its time to set a “Link bus” type timetable, that is running every 10 minutes rather than a scheduled time.
You need a timetable for junction planning purposes.
“…despite an applauding service standard lately” – not sure I’d be applauding this service…
Good spotting, fixed
Amazing how ordinary continous 20% rail ridership growth has become. But it isn’t, it’s extraordinary, and a complete refutation of the naysayers and rail haters that dog the Auckland political and commentary scene.
These people are all being shown to be profoundly wrong about Transport, Auckland, and people’s choices. And while everyone gets things wrong, to persist in that wrongness as the evidence piles up against that view is to fall into perversity and irrationality.
But 20% ridership growth isn’t a good in itself anymore than 20% increases in the number of cars using a motorway would be.
There is no inherent value in buses, trains, cars, segways, or helicopters. What matters is the effectiveness (time) and efficiency (cost) of the system.
Now it may well be that ridership expresses preferences and is thus a useful proxy for the above, but we need to go past these 19th century metrics and ask ourselves…
1. What is the average commute time in Auckland (road vs. PT), and has it come down or not?
10,000 cars at 30 minutes + 20,000 bus travellers at 20 minutes would give an average commute of 23m20s. Is that better than 30,000 cars at 20 minutes?
*timeliness* – real timeliness measures – are more important than growth.
“10,000 cars at 30 minutes + 20,000 bus travellers at 20 minutes would give an average commute of 23m20s. Is that better than 30,000 cars at 20 minutes?”
It certainly is if the cost of getting 30,000 cars to their destinations in 20 minutes [10 minutes sooner] is $1 Billion dollars and the cost of getting the 20,00 bus users to 20 minutes is $100 million.
Then the bus option represents 10 times the value of the car option. And if it cost say $100 million more to get the bus users to 30,000 all at 20 minutes, then again that would still be money well spent compared to the car option.
So, as in all things trade off abound and finding a local “sweet spot” for one mode (just for cars) may not find the overall “sweet spot” across.
I agree entirely, which is why the number of people using a mode of transport is intelligence, not performance.
Not so fast. It’s more complicated than that. You have to try to calculate what journey times would have been if all those people were not using the rail network. Those 2.25m more trips on the rail network, plus the other 12+m already using it, are people not driving. They themselves are not exposed to congestion [they are opting out of that foolish game- but only because they can; because there’s now a decent service], but also they are not contributing to worsening congestion for others. And over time the outcome is likely to lead to an equilibrium between modes, all other things being equal. In other words as the alternative improves and attracts users, then the driving option will be come less dysfunctional and perhaps attract some users back; and so on. However in a fast growing city the practical outcome in fact will be only a less worse outcome on the saturated route, as there are daily new users to replace the converts to the new option.
So in this situation we have to ask where can we add much more capacity quickly and cost effectively. The rail corridor has a huge extra capacity that can be accessed without having to build whole new m’ways or build massive new storage for thousands of cars. It will post CRL be able to carry the equivalent of 18+ m’ways lanes to and through the wider city. Clearly, if riders choose it, this will massively improve the travel times for everyone THAN WHAT THEY WOULD OTHERWISE BE. And that ‘if they choose it’ is important; and means that signs that people are choosing it now is very very significant.
Additionally, you are ignoring the positive externalities of urban PT use and focussing only on one metric. There is so much else that is valuable from people stepping out of their cars to pursue their lives and work in a city, especially if they are using our largely renewably generated electricity to conduct their economic and social transactions: Happiness, Health, Air Quality, less death and injury, etc, etc…
For these reasons and more: such strong growth in uptake of the rail network is absolutely a clear positive for the city.
20% growth in driving would be a disaster; total breakdown, a jump in that negative known as traffic congestion. It would simply exacerbate its inefficiency, ruining your favourite metric; average travel times, and so many others. Whereas rail use growth use makes an existing system more cost effective by increasing throughput, increasing fare revenue, driving financial subsidy down and economic benefit up [especially for non-users].
Does anyone know what sort of signalling system is at use in Auckland?
I mention that only because the Docklands trains in London were a nightmare for several years when they were first put in, and when the system was relatively straightforward (ie only one line, few sets of points etc). They would break down at least once every trip, sometimes more. But now, 2 decades on, and they are achieving fantastic performance. From what I understand, part of this was to do with the signalling system used – they changed to a “rolling block” signal system, which seems to have made things a lot better. But it sped up time at junctions, and both leaving and arriving at platforms. Cost millions to change it of course, but that’s another story….
Back on the trains this week after a few weeks of having to take alternative transport. Have been lucky so far with only some late running and slow trains on the western line, although yesterday morning the Britomart hold-up was staffing issues. That shouldn’t be happening as much.
Sounds like I’ve been lucky compared to some folks experience. Bring on the full rollout of electrics!
ETCS level 1. Rolling block works well on a self contained system like the DLR but it is very expensive for a combined freight/metro railway
I apologise if this issue has been explained before, but I have to ask why AT measure the service delivery for bus and ferry on the departure time, whilst on the trains it is based on the arrival time at the final destination. This makes any comparison of the performance between the train and bus/ferry spurious. To be able to make a fair comparison of the quality of the services, I would suggest that it be standardised on whether the service completed the planned journey and the arrival time at the final destination. I suspect that this would show a much lower performance on services outside of the RTN.
i would like my performance at work be measured by what time i walk in to the office, regardless by how long I’m going to stay there
I’d like my projects measured by their start time, not their end date!
I assume it is because the train operator controls the traffic (on the rails) but a bus operator does not control the traffic (on the roads).
If they were to measure reliability at the end point, assuming the bus started on time, any delay is AT fault (due to not enough bus lanes etc), not the bus companies.
Bus operators are impacted however as most route operate back to back. So if you are late arriving, then your next trip will be late departing (there is obviously some buffer as anyone who has sat on a Link bus at Britomart knows).
good to see HOP uptake continuing to grow. Need to drive it up towards 90% and then move to no change given onboard as the precursor to driving the final nail into the paper ticket coffin.
Do you move to no change or higher HOP discount first. No change is very mean when all you have is a $20 and your HOP card is locked because your credit card expired.