April-15 Patronage

Another month and another good patronage result from Auckland Transport – particularly for rail. Patronage in April is naturally down on the madness that is March due to the combination of a 30 day month, ANZAC day, Easter and School Holidays/Uni holidays. This year was no different although there ended up being the same number of working days as April 2014. Overall patronage for the month edged up 3.7% compared to April 2014 however there is quite some variation between the modes.


2015-04 - Patronage Table

2015-04 - Total Patronage Chart

The real focal point – as it has been for many months now – has been the stellar growth in rail patronage. In April it hasn’t disappointed, up almost 16% compared to April 2014 and up 22% annually and even more for both measures if normalised to take account of the differences e.g. events. To put things in perspective, 12 months ago the annual patronage on the rail network was just under 11.1 million trips, now it’s over 13.5 million. That means it remains well on track to exceed the government’s patronage targets for the CRL some time during 2017/18. It’s also worth noting that AT have now upped their projection for this financial year (end of June) to suggest that we’ll reach 13.8 million trips


2015-04 - Rail vs CRL target

In some ways I think AT are lucky that achieved the results they did given that operational performance was so bad achieving just 68.4% of services arriving within 5 minutes of schedule.

2015-04 - Rail Performance

With buses the Northern Express continues to perform well and was up over 8.5% for the month and 17% for the year once again showing it’s the Rapid Transport Network is where the most growth is happening. Other buses were actually down slightly although a reason for this isn’t given.

Ferries have had surprisingly strong growth of late and were up almost 15% for the month. AT suggest that a large part of the growth has some from the new Explore ferries.

2015-04 - Ferry

Lastly a quick update to my post last week about train costs. In it I included a chart showing that subsidies per passenger km were starting to decline on the rail network which is a good thing.  The stats for this month show once again subsidies are reducing which will be the result of more and more electric trains coming in to service. In a few months I’d expect that line to be even lower too.

PT Subsidy per Passenger km

Why is Auckland’s rail system so expensive to operate?

On Monday the NZ Herald ran an article about the cost comparisons between operating Auckland and Wellington’s rail services. Based on the numbers in the article, it doesn’t tell a particularly pretty story:

Aucklanders facing a new transport rate are paying far too much to subsidise the city’s trains, says councillor Mike Lee, a long-time champion of rail.

Mr Lee, the council’s infrastructure chairman, has calculated that Auckland trains are costing ratepayers and the Government four times more to run than those in Wellington for each kilometre travelled by passengers.

He has worked out from figures published for each rail operation that the subsidy paid for every “passenger kilometre” travelled on Auckland trains last financial year was 65c, compared with just 16c in the capital…

…He said subsidies covered $107.2 million of Auckland’s total rail operating costs of $139.3 million last year, when 11.4 million passenger trips were made on trains.

That compared with just $42.7 million of subsidies towards the $85.09 million cost of providing 11.6 million passenger trips in Wellington.

Although the Government covered $63.92 million of Auckland’s bill, ratepayers forked out $43.3 million.

Auckland rail passengers also paid more for each kilometre they travelled, 18.5c against 16c in Wellington.

An Auckland Transport spokesman queried Mr Lee’s finding that fares covered only 22 per cent of rail operating costs, and sent the Herald a report prepared for the council by the organisation’s chief financial officer, Richard Morris.

But the report, despite pointing to difficulties in comparing the Auckland and Wellington rail operations, did not directly challenge Mr Lee’s figures.

A year ago Auckland Transport staff wrote a report breaking down the costs for each system which is the one referenced above. It split out the costs for the 2011/12 and 2012/13 years for each system. I haven’t seen 2013/14 figures and we are still in the 2014/15 year.

Overall it can be difficult to pinpoint the exact differences and while not definitive AT have identified some of reasons why they think Auckland’s system is more expensive to operate. This includes:

  • Auckland is running a mixed fleet with expensive to fuel and maintain diesel trains. The locomotives that haul the SA sets are also leased from Kiwirail. I’ve been told in the past that fuel and maintenance costs will roughly halve once all services are electric.
  • The mixed fleet reduces driver roster flexibility (more overtime to cover issues) and increases training costs. This should improve once all fleet electric
  • Auckland’s trains are slower for a variety of reasons. This increases the time it takes for each trip and therefore staffing costs for similar length journeys. We’re yet to see if electrification will improve this.
  • Auckland costs include all of Transdev’s wages and salary’s while in Wellington it’s believed some of Kiwirail’s overheads are likely shared or covered by the fact those roles exist for the rest of the business – both Auckland and Wellington are currently tendering for rail services as they moving to the same contract model this should hopefully see costs move closer together.
  • Auckland has a number of stations which are more expensive to run – e.g. it appears that Britomart alone costs more to operate than all the Wellington stations combined. Auckland also supposedly has more security monitoring. Auckland’s costs will remain higher in this area.
  • I believe that overall Auckland now runs more services than Wellington does – and this will have increased even further since December.

AKL vs WLG rail costs

Since the data that report patronage has gone crazy and is up about 3.5 million trips. We’ve also started to see electric trains on parts of the network and the remaining lines are due to go electric soon. We’ll have to wait to find the final costs for this year however this chart from AT’s monthly indicators in March shows the subsidy per passenger km and it does appear that things are starting to head in the right direction.

AKL Subsidy per PAX km

The minutes for last week’s Budget Committee meeting aren’t yet online, but it seems like the Committee passed a resolution to request an investigation into the issue and report back on options for cost savings in the next round of annual planning. This seems like a pretty sensible approach as we wouldn’t want savings to be achieved by cutting (or not increasing) service levels, but rather through a better rail contract and continued patronage (and therefore revenue) growth.

Western Line Cancellation Reasons

As regular readers or train users will know, the service offered recently hasn’t exactly been great with many cancellations and delays. When it comes to cancellations the Western Line has been hit the hardest and in March 6% of all services on the line failed to reach their destination (reliability) – that’s 181 services all up. Note, this is a different measure from whether they arrived on time (punctuality) and of the services that did reach their destination only 73% were within the 5 minute requirement. The other lines weren’t too much better with them combined averaging just 95.5% punctuality and 81.4% reliability.

Radio NZ has picked up on the issues with the Western Line and managed to obtain a list of all 181 reasons that trains were cancelled for.

Radio New Zealand has obtained from Auckland Transport, a “raw” list of reasons for March’s cancellations on the most troubled line, to the west. There’s a wide variety of human and technical reasons in this selection from the 181 :

  • Driver got sick at work
  • Late running train due to passenger loadings (previous train cancelled)
  • Driver late for train
  • Service terminated at Avondale due to door fault
  • Roster planning – shift not covered
  • Heavy tagging of service while waiting at signal
  • Trespasser in Britomart tunnel delayed departure
  • Delayed due to loading wheelchair-bound passenger – cancelled to recover timetable
  • Roster planning – no driver
  • Trespasser riding on outside of train
  • Faulty driver seat
  • Unfilled shift – compassionate leave
  • Rolling stock short supply, unable to cover service
  • Train fault – headlight not working

The agency’s general manager, Mark Lambert, said Auckland Transport knew it had to do better.

“The performance in the last few weeks and months has been unacceptable,” he said. “We don’t want to be there, and we’re looking at everything we can in terms of immediate improvements, but also how we can bring forward initiatives to improve things.”

Mr Lambert said rail services had been hit by the phasing out of the old, unreliable diesel trains. As the maintenance contract at KiwiRail wound down, staff had begun leaving, and the standard of maintenance had fallen.

The progressive introduction of the new Spanish-built electric trains had also been affected by minor bedding-in problems, all of which were being sorted, he said.

This is only a snapshot of the full 181 cancellations but the thing that really stands out to me are the number of mentioned which are due not to technical issues but roster ones such as no driver being available to drive the train. That could be from drivers being delayed on one service missing their next one but it’s not the only reason with the number of drivers available to run services also being impacted by the roll out of the electric trains. It’s this reason why AT have also stated they want to get all services converted to electric operations by the end of July.

There’s also this report on the issue from Radio NZ which includes some vox pops from train users. One thing that surprised me was how positive most seemed to be with comments like issues happen in any transport system, that they are looking forward to the electrics and how good it is that so many people are using PT. I’m not sure what the reason for this is but perhaps most users are much more forgiving in these issues than we and most readers have expressed recently.

Or listen here

Western Line Padding Part 2

A couple of days ago I wrote Auckland Transport padding out the Western Line timetable. Today the Herald picked up on the story. AT have sent me a response to the story which is in full below.

AT is committed to running reliable peak services and achieving this through the introduction of a world class all-electric fleet.

In order to address the recent decrease in reliability and performance, the decommissioning of the diesel fleet has been brought forward to July from a planned August / September date. This will remove one variable of on-time performance – aging diesel train breakdowns that have seen a spike in recent months.

As with any timetable change, the most recent being on 8 December with a +22% increase in services, tweaks are required following introduction. The 3 minute earlier departure times on the Western Line are needed to support the 8 December increased services while diesel and electric trains share the network, to ensure time slots at Newmarket junction and Britomart are met.

As part of the original electric train rollout plan and in accordance with that plan, run-times of the electric trains on the Western Line are being tested and optimised currently prior to introduction in July as part of the existing timetable. Continuous improvements in run times, signalling and track testing to optimise electric train run times are a priority and further run-time testing and optimisation will follow the bedding in of full electric train operations from July, leading to more timetable improvements later in 2015. This work involves looking at reducing dwell times and optimising the ETCS (European Train Control System) safety feature on the electric trains.

AT is committed to providing the peak 10 minute frequency on the Western Line as early as possible and will be dependent on progress over coming months.

The historic run times relate to when run times were increased to accommodate the SA train sets. To place this in context, the improvements on the rail network since 2003:

Train Changes 2003-2015
I think it would be good for AT to actually specify exactly what the impediments are to faster and more frequent services and what steps they are taking i.e. how much do they need to reduce dwell times, what are the changes that need to occur for 10 minute frequencies etc.

As often happens when AT release data like above it manages to highlight a couple of other important points. In this case that a few things I noticed:

  • Patronage has increased well above the increase in the number of services per week. This is a good things and shows suggests we’re not only getting better services but also more utilisation out of them.
  • At 30 arrivals at Britomart over a two hour period it highlights just how much more capacity the CRL has the potential to contribute. Although first a quick count suggests the actual number of AM Peak (7am-9am) arrivals is 36 trains which would go to 40 if the western line moved to 10 minute frequencies. My understanding is the CRL allows for up to 48 trains an hour (24 trains per direction) so that’s up to 96 over a two hour window. In other words the CRL more than doubles the capacity of the rail network.
  • Back in early March AT said that on average weekdays in February there were 51,500 trips. For March that has increased to almost 60,000 per weekday

Western Line Timetable padding

Regular users of the western line will know that in the services have been struggling in recent months and that’s even when trains aren’t breaking down. It’s especially noticeable in the morning peak when and while there’s been a little bit of relief during the school holidays, with schools back today I’m expecting services will have been chocka again (I say that having written this post on the weekend).

Over the last few years patronage has been growing strongly and in the last 12 months trips up over 16%. That’s increase is not quite as high as the lines which have been electrified but is still a very significant rise. This is even more so considering that there hasn’t been an increase in peak frequency since 2008 despite during that time double tracking has been completed. The timetable changes that have occurred over the years have extended the length of the peak and improved off peak frequencies which only a few years ago improved to half hourly.

2015-03 Western Line Patronage

The yearlong bump was due to the Rugby World Cup


One little side note is that based on a patronage per service measure the western line carries about 60% more passengers on each train than what is seen on the other lines. I suspect that this is in part due to the Western line having a higher number of trips to destinations along the line rather than just to Britomart. This is kind of shown in the great visualisations put together by Aaron Schiff

Aaron Schiff Visualisation - away from Britomart


While increasing patronage is a good thing, at peak times the lack of improvements in capacity – with the exception of a few extra carriages – has had some negative effects. Now trains are regularly crowded with some passengers sometimes standing from as far as Glen Eden or even Henderson. Those from even closer in such as Kingsland sometimes aren’t able to get on at all – and before anyone mentions it, the buses going past on New North Rd or Sandringham Rd are also full. This is obviously bad for customers but it’s also bad for operations. It means more time needs to be allowed at each station for people to push their way to the doors or find a space to squeeze on while they wait for people to reluctantly move away from the doors and down the aisles. This slows the trains down and therefore increases chance that the service will run late.

Poor punctuality frustrates people and in some cases will put them off using the service altogether. In recent months punctuality has declined and in February it was less than 80%. In other words more than 1 in every 5 trains will run late. The performance results for February are below

2014-02 - AKL - Train Performance

When delays are occurring due to busy services – or “heavy customer demand” as Transdev would say, there aren’t always that many solutions. The most obvious would be to increase capacity but that isn’t frequently something that can happen quickly. I’d also point out that it was promised that the western line would move to 10 minute services in the peaks in 2010 when double tracking was finished but numerous excuses have continued to be created as to why that can’t happen. The latest excuse is the Sarawia St level crossing and I suspect that once that’s sorted I’m sure they will find another one such as capacity constraints at Britomart or Newmarket or perhaps the CRL works.

A shorter term solution is to simply extend the timetable by padding it out further and it appears that’s exactly what AT have done. This poster started appearing on trains last week.

Western Line Timetable increase poster

The new timetable is here and as it says, travel times are extended by up to three minutes. Below is a quick comparison of travel times between the December 2014 and April 2015 timetables and as you can see a minute has been added between:

  • Sturges Rd and Henderson,
  • New Lynn and Avondale
  • Mt Albert and Baldwin Ave

Western Line Timetable increase num


I have a few issues with this approach. While the timetable might more accurately reflect the times that customers will see it does so by making the services slower and therefore less attractive to potential new customers. It also improves AT/Transev’s performance results by simply shifting out the measurement.

Of course this isn’t the first time the timetable has been padded out. The graph below shows the travel times from Swanson of a few of the different timetables that there have been on the Western Line since Britomart opened. In total the new timetable is up to 8 minutes slower than was achieved back in 2003.

Western Line Timetable time changes

Despite what’s been mentioned above the timing of the change suggests there’s possibly another reason for the timetable increase and it’s one I certainly hope I’m wrong about. AT have said recently that they now intend to have electric trains running on all services by August. From what we’ve seen on other lines that should mean we start seeing electric trains running off peak shortly and given that it would seem odd to increase the timetable now. That is unless the testing of the trains has revealed that they’ll be even slower than the old diesel units they’re replacing. It would be seriously disappointing if this were the case as times should be going the other way. As I said, I really hope I’m wrong (and that AT clarify this).

What’s with ‘The Void’?

Could Auckland have something like this running on a couple of major city routes before this decade is out? The AT board is to decide later this month how to proceed with its Light Rail plan and with what sort of pace. Everybody it seems loves trams, but why now and why there? What problem are they addressing? In a follow-up post I will discuss the financial side of the proposal.

CAF Urbos Light Rail for Utrecht

CAF Urbos Tram recently ordered by Utrecht

First of all lets have a look at Auckland’s situation in general terms. Auckland is at a particular but quite standard point in its urban development: 1.5 million people is a city. The fifth biggest in Australasia; behind Sydney, Melbourne, Brisbane, and Perth. But on the location with the tightest natural constraints of the group; squeezed by harbours, coasts, ranges, and productive and/or swampy farmland, it shares the highest density of the group with Sydney in its built up area. And is growing strongly. It also has the poorest Transit network of the group and consequently the lowest per capita Transit modeshare [although the fastest improving one].

So these three factors scale, growth, and density are all combining to create some serious pressure points that require fresh solutions especially on existing transport routes, and particularly on the harbour constrained city isthmus.

This pressure is on all transport infrastructure, at every scale from footpaths [eg Central City, Ponsonby Road]; the desire for safe cycling routes; on the buses, trains, and ferries; to road space for trucks and tradies, and of course road and street space for private vehicle users. Transit demand in particular is going through the roof and this is way ahead of population growth and traffic demand growth, especially at the higher quality Rapid Transit type of service where growth over the last year has been at an atsonishing 20%.

This is to be expected in a city of Auckland’s current state as Transit demand typically accelerates in advance of population in cities of a certain size, because of the universal laws of urban spatial geometry, as explained here by Jarrett Walker;

This problem is mathematically inevitable.  

As cities grow, and especially as they grow denser, the need for transit generally rises faster than population, at least in the range of densities that is common in North America.  This is completely obvious if you think about it, and I stepped through it in more detail in Chapter 10 of Human Transit.  In brief: Suppose a particular square mile of the city doubles in population.  Transit demand would double because there are twice as many people for whom transit is competing.  But independently of that, if density is higher, each person is likely to find transit more useful, because (a) density creates more disincentives to driving and car ownership while (b) density makes it easier for transit agencies to provide abundant and useful service.   Those two separate impacts of density on transit, multiplied together, mean that transit demand is rising faster than population. Again, go to my book for a more extended and thorough argument.

And that this means that the infrastructure needs of our growing city is likely to be ‘lumpy’. Big long lasting kit that is costly and disruptive to build become suddenly urgent:

As transit demand grows in a growing city, it hits crisis points where the current infrastructure is no longer adequate to serve the number of people who want to travel.  Several major subway projects now in development are the result of transit’s overwhelming success using buses.  I’m thinking, for example, of Second Avenue in New York, Eglinton in Toronto, Wilshire in Los Angeles,  Broadway in Vancouver, and Stockton-Columbus in San Francisco.

Broadway, for example, has local buses running alongside express buses, coming as often as every 3 minutes peak hours, and they are all packed.  In that situation, you’ve done just about everything you can with buses, so the case for a rail project is pretty airtight.   In all of the cases I mention, the rail project usually has to be a subway, because once an area is that dense, it is difficult to commandeer enough surface street space, and we tend to have strong aesthetic objections to elevated lines in these contexts.

As driving amenity is very mature in Auckland there is very little opportunity to add significant driving capacity to streets and roads to much of the city at any kind of cost, and certainly not without a great deal of destruction of the built environment. This has long been the case so in a desire to solve capacity and access issues with a driving only solution we did spend the second half of the last century bulldozing large swathes of the Victorian inner suburbs into to make room for this spatially very hungry mode. This solution is no longer desirable nor workable. Below is an image showing the scar of the Dominion Rd extension citywards and the still extant Dom/New North Rd flyover. These were to be the beginning of a motorway parallel to Dominion rd to ‘open up’ or ‘access’ the old isthmus suburbs.

1963, Dominion Rd flyover in the foreground

1963, Dominion Rd flyover in the foreground

Where we can’t nor want to build ever wider roads we can of course add that needed capacity though the higher capacity and spatial efficiency of Transit. Most easily with buses and bus lanes. There are also potential significant gains to made at the margins by incentivising the Active modes with safe routes especially to Transit stations and schools and other local amenity.

However as Jarrett Walker describes above there comes a point where buses, through their own success, cannot handle the demand as the number of vehicles required start to become both less efficient and more disruptive than is desirable. At this point demand can only be met with higher capacity systems with clearer right of ways. Such systems require expensive permanent infrastructure and are never undertaken lightly.  The CRL, being underground, clearly fits this definition and is due to begin in earnest in the new year. And although the physical work and all of the disruption of the CRL build occurs in the Centre City, the capacity and frequency improvements are to the entire rail network, and therefore much of the city: West, East, and South.

But not everywhere. Not the North Shore, not the North West, and not in ‘the Void’, as AT call it, the isthmus area between the Western and Southern Lines. Shown below in purple with the post CRL Rapid Transit Network. This area has a fairly solid and quite consistent density, housing about the same number of people as West Auckland, around 150,000. Note also the South Eastern Busway [AMETI] plugging directly into Panmure is very much a kind of rail extension for the Transit-less South-East, as is the Manukau spur further south.

RTN Void

The Void

These three major areas will still be relying on buses. The CRL, New Bus Network, and Integrated Fares will enable and incentivise more bus-to-train transfers that expand the reach of the core rail network and that this will help limit the numbers of buses going on all the way to the city. But this is primarily for the South, South-East, and West of New Lynn, there will still be an ever increasing number of  buses with from the remaining areas converging on the City Centre. AT calculates that we need to act now to cut the bus numbers from at least one of these major sources to leave room for growth from the others, and all the other users and uses of city streets. [More detail on this in Matt’s previous post, here].

The North Western is currently getting more bus priority with the motorway widening, and hopefully proper stations at Pt Chevalier, Te Atatu, and Lincoln Rd [although NZTA and/or the government are showing little urgency with this aspect of the route]. Also priority improvements to Great North Rd and further west too. The North Shore is the only one of the three with a Rapid Transit system [which also should be being extended now], and while there is still plenty of capacity on the Busway itself, like the other routes these buses are constrained once in the city. This leaves the very full and frequent ‘Void’ bus routes as the ones to address with another solution first.

So essentially LRT for this area has been selected because of the need:

  • for higher capacity and efficiency on core Isthmus bus routes
  • to reduce bus numbers on these routes and especially in the central city
  • adds Queen St as an additional high capacity North-South city route
  • for extra capacity both before and after CRL is operational
  • to address Auckland Plan air quality, carbon emissions, and resilience aims
  • to enable major public realm improvements along routes, especially Queen St

and possibly because:

  • it may be able to be financed as a PPP so helps smooth out the capital cost of building both projects [more on this in a follow up post]
Above is a schematic from AT showing the two proposed LRT branches. The western one leading to Queen St via Ian Mackinnon Drive from Dominion and Sandringham Roads, the eastern one down Symonds St from Manukau and Mt Eden Roads, some or all routes connecting through to Wynyard Quarter. More description in this post by Matt.
It is worth noting that this area, The Void, gets its very successful and desirable urban form from this very technology; these are our premier ‘tram-built’ suburbs. With all the key features; an efficient grid street pattern, mixed use higher density on the tram corridors, excellent walking shortcuts and desire lines. So what the old tram made the new tram can serve well too.
Auckland Isthmus tramlines

Auckland Isthmus tramlines

With all door boarding and greater capacity LRT will speed more people along these routes with fewer vehicles and lower staffing numbers. Frequency will actually drop from the current peak every 3 minutes down to 5 or 7 minutes [I’m guessing]. This along with the narrower footprint required by LRT is a big plus for other users of the corridor. But the huge gain in travel time comes from improvement to the right of way and intersection priority that can be delivered with the system. Stops are presumably to be at intersections, instead of midblock as buses are, so the passenger pick-ups are coordinated with traffic lights.
But best of all for this writer is that LRT is a tool to drive enormous and permanent place uplift. The removal of cars and buses from Queen St, improvements to New North and Dominion Rds, hopefully including that intersection itself, a fantastic new Dominion road with the potential for real uplift to premier status.  It will spur the redevelopment of the mixed uses zone all along Dominion Rd. This is real place quality transport investment. And all of course while moving thousands and thousands of people totally pollution free and with our own mostly renewably generated electrons. Breathing in the Queen St valley will become a fresh new experience.
Light Rail in Queen St 1 - Nilut
We all look forward to hearing the proposed details of the routes and of course the financials. I will follow up this post with my understanding of the thinking on this next.
Light Rail in Queen St 2 - Nilut
Finally it is very good to see that there is no dispute over the necessary solutions to Auckland’s access and place quality issues, just the details and timing. Auckland Transport’s map above is pretty much the same as our solution in the CFN. We are delighted that AT are planning for four light rail routes were we proposed one.
There are of course plenty of debates to had about further extensions to the Transit networks that this proposal invites; LRT in a tunnel from Wynyard to Onewa, Akoranga, and Takapuna? Then up the Busway? From Onehunga to through Mangere to the Airport? Along Grey Lynn’s apartment lined Great North Road, to Pt Chevalier, and the North Western? Panmure, Pakuranga, Botany, Manukau City Airport? Which of these need to be true grade separate Rapid Transit and for which are bus lanes or busways a more cost effective option? Are their others that would be better suited to extending the rail network? Is there enough density elsewhere in the city to justify other LRT routes?
CFN 2030 + Light Metro

Reliability is key to patronage

The other day after another morning hearing on social media about trains being cancelled and delayed (on that day it was the Eastern Line having problems) Next City just happened to tweet this post from their archives about the issue of unreliability.

It isn’t groundbreaking to find that public transportation users would prefer their bus and train service to keep to a reliable schedule. More significant than that, though, is the idea that unreliable service trumps other factors, including commute length, when riders consider ditching public transit altogether.

A November study from researchers at the University of California, Berkeley found that inconsistent service — whether it’s lateness, inaccurate arrival info or mid-travel delays — is the number-one factor that encourages people to stop using public transit. Chief among the issues that fall under the “reliability” umbrella are long gaps between transfers and wait times exceeding 10 minutes.

Focusing on San Francisco, researchers asked what factors public transit users considered most important, finding that most would rather see their buses or trains arrive more frequently than become less crowded or offer fewer seats.

Notably, riders tended to stress that only delays perceived to be the fault of a given transportation agency would compel them to give up on transit. So, if a connecting train doesn’t show at the right time, it’s more of a deal-breaker than, say, delays due to emergencies or, in the case of buses, traffic jams.

This means that transit agencies, if they want to retain riders, should consider running buses and trains more frequently, even if it means making the vehicles or train cars smaller. Also, communication never hurts. Few things are more grating than getting stuck in a subway tunnel without knowing the reason — letting passengers know that a problem is out of the driver’s hands will make them less inclined to misplace the blame.

There’s more on it in this article that was linked in the post.

What this helps to highlight is just how important it is that services are reliable. If they’re not – like Auckland’s train services have been lately – the end result is going to be less people using them. The delays are primarily being caused by the old clapped out diesel trains breaking down as a result of maintenance being scaled back due to their pending retirement. It’s not the only reason though as there have also been issues with freight trains and staff shortages which can partly be blamed on events like the Cricket World Cup which required more and later services to be run.


In one piece of hopefully good news, it seems Auckland Transport are realising this with the Herald reporting that they’re trying to speed up the roll out of the new electric trains in a bid to sort out the issues that stem from the old trains.

Auckland Transport is trying to accelerate its rollout of electric trains, after continuing disruptions from diesel breakdowns.

Although August remains the official target for completing the rollout by extending electric trains to the western line, the organisation’s board was told yesterday of efforts to bring the date forward.

Public transport group manager Mark Lambert said that would allow an earlier retirement of much of the increasingly troublesome diesel fleet, although 10 multiple-unit ADL trains would be kept and eventually refurbished for shuttle duties between Pukekohe and Papakura.


Mr Lambert said there had been a high rate of diesel train breakdowns in January and February, which tended to cause greater disruption than previously, because of a higher frequency of services on the rail network.

“When it happens it has a bigger impact because of higher frequency,” he said.

Train punctuality was poor in February (below) and I expect it will be even worse in March given what we’ve been seeing.

2014-02 - AKL - Train Performance

As reliability also impacts on patronage heavily it’s probably also been a factor in the massive growth that’s been seen over the 18 months or so.

Rail Network Reliability Feb 15

Getting the EMU’s rolled out sooner – assuming it’s possible – should be a big boost not only for reliability but also for capacity which in many places is extremely stretched. Let’s hope AT can do this before they lose too many customers.

Why are trains failing so much?

With March Madness in full swing it seems that Auckland Transport and Transdev have taken a new approach to dealing with the high demand, driving customers away through rubbish performance.

Train users from across the region – but it seems particularly West Auckland – have been suffering after what feels like daily issues that are putting even more pressure on already stretched services. People can accept one or two issues but when they become an almost daily occurrence the only result will be less people using trains.

Here are a just couple of examples from the last week or so of what we’ve seen happening from AT’s text message service. It is by no means an exhaustive list of issues that have occurred and is only on the western line. There have also been issues on other lines too (the am times are based on services at my local station – Sturges Rd, pm times based on Britomart).

13 March

7:43am train running at reduced seating capacity – a later text states it was full from Avondale

5:36pm train running at reduced capacity

16 March

4:36pm train delayed 10 minutes due to an earlier train fault

5:06 train delayed 20 minutes due to an earlier train fault

5:36 train delayed 15 minutes due to an earlier train fault

5:50 train delayed 25 minutes due to an earlier train fault

17 March

4:36pm train cancelled from Kingsland due to a train fault

4:50pm train delayed 10 minutes due to an earlier train fault

19 March

6:59am train cancelled due to vandalism

5:20pm train running at reduced seating capacity

20 March

7:43am train delayed 10 minutes due to a train crew matter

5:50pm train delayed by 10 minutes due to an earlier train fault

23 March

7:43am train cancelled due to an earlier train fault

The reason these are so bad particularly on the western line is any cancelled train means the next train is at least 15 minutes away. Depending on what station you use that might be only the delay as trains that have been so full thanks to the surge in patronage mean that following services simply can’t cope. As such people from inner stations have been unable to even get on some trains meaning potentially they could be experiencing a 45 minute or longer wait. Even if you’re lucky enough to get on a train in these situations it is going to be a cramped affair.

Of course this wouldn’t be quite so bad if frequency was every 10 minutes like first promised for the western line in 2010 but AT have managed to find a constant stream of excuses as to why it can’t happen.

So what’s causing these delays and cancellations? My understanding – although I’m happy to be corrected by AT and/or Transdev is that it’s a combination of few things. Maintenance is understandably being reduced on the aging diesel trains ahead of the roll out of the electrics that is leading to more of them failing. They also are extremely stretched with their staff numbers and don’t really have enough drivers to run the timetable properly.

At the end of the day it doesn’t really matter what the reasons are for the delays, what’s clear is that all those involved need to get this issue addressed extremely quickly otherwise the great patronage gains that we’ve seen in the last 1½ years will plateau again or even fall away. This is because the two most important aspects of any PT service regardless of mode or what city it is in are always frequency and reliability

Panmure Station Revisited

Train Bus Interchange. Looked to me like was working pretty sweetly. Quite a bit of Kiss’n’Ride going on on the northern side, car drop off, as you’d expect for a reasonably far enough out station in such an auto-dependent city. And, rather like New Lynn, this station feels somewhat stranded by roads and not anything like the intensity of land use we all expect to see develop over time.


But of course those roads bring the buses right to the front door; quite a lot of people seem to be transferring to the trains rather than staying on the bus all the way to the city centre, and Howick and Eastern looked to be doing a good trade to and from the station. It is interesting that H&E have just announced they are buying 15 new double deckers, all with wifi and charging points. It looks like the quality of the new trains has started an quality of service race among providers, along with providing the core of the lift in ridership enabling this sort of investment and upgrade; win win win.


Looking forward to the next Interchanges at Otahuhu and Manukau that are funded to start this year. However the really spectacular upgrade for SE Auckland will be the Bus Rapid Transit part of AMETI which will connect this station with Botany, Pakuranga, and hopefully Highland Park with bus priority [construction start 2017]. Won’t be too long before we have new and much better options for getting around our city.



A Parnell Rail Station by June?

One of the projects that is sitting in the pile awaiting funding is the Parnell Train Station. The project has been one that seems to always be just around the corner. The tracks in the area were lowered in 2011/12 in one of largest Christmas shutdown’s we’ve seen to enable the station to be built but that never happened and constant delays have ensued. One of the major problems is the cost which is estimated at close to $20 million.

That’s definitely a lot of money however with its proximity to Parnell, the rapidly developing area around the old Carlaw Park and with it being the closest station to much of the University it has the potential to be one of the busiest stations on the network. The plans would see the old Newmarket station that is currently in storage moved to the site and restored with the intention of tying in with the mainline steam site.

Parnell Station

There are a number of other images in this post. With funding currently dependant on the outcome of the LTP discussions I had assumed it the project had been placed in stasis. However it now appears that might not be the case.

Deep in the agenda for the council’s Parks, Recreation and Sport Committee next Tuesday is an item about Auckland Transport seeking approval for works within the Auckland Domain in relation to the Parnell Station. The approval is needed because some of the construction works would happen outside the rail corridor and in the Domain itself. It says that AT seeking approval for the works on what will be stage one of a revised version of the station and consists of just the platforms and paths. Crucially it says that AT are wanting this part of the project done very quickly with construction complete by June this year. This sounds a lot like they’re trying to use up some remaining budget.

My first reaction was that this seems like a very proactive move from AT, rather than waiting for the funds needed for a large redevelopment, get in there and at least get something underway and working. However upon looking at what’s just what’s planned – or in this case what’s not – I’m now less convinced. The works to be constructed within the rail designation are:

  • The proposed platforms are to be constructed adjacent to the existing Mainline Steam building, as shown in the attached plan. Both platforms are 5.0 metres wide and approximately 16.0 metres long. [presumably a typo and they mean platforms 160m long]
  • New pedestrian pathways from the existing underpass to the platforms. CCTV operation and lighting of the platform.
  • Platform seating, service points, station signs and necessary accessibility considerations (stairs/ramps).

So missing from this list we obviously have the station building – which I’m not that keen on anyway – but more importantly it seems no shelter at all. In addition there will be no pedestrian over-bridge, something I’ll cover shortly. Below is the current plan.

Revised Station 1 - Feb 15

Other than the missing shelter and pedestrian bridge the other thing that surprises me – but perhaps shouldn’t – is that it seems there’ll be a substantial area for car drop off and turn around. Surely the last thing we want to do is encourage people to be driven to the station clogging up those narrow and steep streets.

Along with the shelter, perhaps the most serious issue is the lack of the pedestrian over-bridge between the platforms. Without it means the only way for someone coming from the western side of the station to access the eastern (southbound) platforms is via the existing underpass over 100 to the south of the platforms. That means all up it’s around a 400-500m detour. That alone will put a lot of people off using the station. In fact I think it’s so serious it could backfire on AT and playing right into the hands of the anti-CRL brigade who will hold it up and say the same thing will happen with that project.

I should add I’ve long been lukewarm on this project as I’ve thought the station has been placed in the wrong location. The site was chosen for its proximity to the Mainline Steam site – after some desired a sort of heritage precinct – it’s straight line proximity to Parnell – ignoring the steep and narrow streets – and it was also argued that it would be a museum station again ignoring the steep hill between the two locations.

Instead I’ve long thought it should be on the outcrop slightly further north that’s now used as a carpark. It would have provided easy connections to Heather St for an easy walk to the Parnell main street, to Parnell Rise for good connections to Link buses, better connections to the current and future development in the area. Further a path along the rail line would have still provided very easy access to the Mainline Steam site.

Parnell Station Alternate Location sketch

Unfortunately it’s probably a bit too late for this as the cost in both money and disruption to change now would likely be too great. I can probably live with the current planned location and have no issue with AT looking at where it can cut costs – such as dealing with the station building later. However it the plan is for the station to be so basic as to exclude shelter and easy access between the platforms then I have to ask why bother at all.