In two previous posts (here and here) I have discussed some factors that seem likely to impact on the development of transport technologies. To sum up, these posts discussed:

  • External factors that impact the development of new technologies, including land use patterns, public policy settings, and demographics; and
  • Economic factors that impact the uptake of different transport technologies, including economies of density, fixed/variable costs, and complements versus substitutes.

This post will now consider some emerging transport technologies in more detail. These technologies seem likely to make us much better off in the next couple of years, and are worth getting cautiously excited about.

1. Information technologies – The brain in the palm of your hand

Hypothesis #1: Improvements in IT technology will make it easier for people to access the information they need to make good choices. When presented with this information, people will travel by the best transport mode for that particular journey, rather than the one that is easiest. This will reduce the dominant role played by private vehicles in our current transport system, especially in denser cities where a variety of transport options are available. By reducing the complexity and cost of getting around, these IT technologies will also increase the attractiveness of dense cities as places to live.

Some of you may be reading this post on your phone. On the same phone you may also have access to apps such as Google Maps, Uber, AirBnB, and Tinder. With these apps you can check bus/train timetables, book a (cheap) taxi, find a cool place to stay, and arrange a hot date. They help you access the goods and services you want in an efficient manner.

The long-run impact of these kinds of technologies cannot be overstated. In barely 20 years, for example, online dating has completely transformed how we meet the people we love, as illustrated in the figure below (source). Of course, more traditional forms of romantic encounters remain important, but they are *less* important than they were previously. In this way IT has expanded our options and enabled us to do something we were already doing, just more effectively. They function at their best in areas of high density.


NB: 1) This data only goes up to 2009, i.e. it excludes the more recent advent of Tinder and 2) the % of same-sex couples who meet via online dating is higher still.

Google Maps, Tinder, and Uber are old (albeit good) news. What’s the next big transport IT to emerge?

I think it has to be ride-sharing. Uber themselves are operating a shared version of their service in some cities. And they’re not the only ones: The graph below, for example, shows the growth of “Bla bla car” in Europe from 2009 to 2014. Year-on-year growth of 100-200% has enabled Bla bla car to raise hundreds of millions in venture capital to fund their ongoing expansion.


Bla Bla car works as follows: 1) Go to a website; 2) Enter where you are travelling to/from, and 3) you get a list of people who are driving that journey, their time of departure, and the cost. It costs about 5 Euro to get from my house in Amsterdam to Rotterdam, which is a 75km journey at 6 Euro *cents* per km, i.e. cheap as chips. Many drivers are happy to carry pets, and some are commuting regularly, so you can negotiate arrangements for a more regular journey.

Now there’s been lots of (failed) attempts at building ride-share platforms in the past, so why am I so confident that something like Uber’s shared service or Bla bla car will take off soon? Two main factors make me think things are different now. The first is psychological: In the last decade or so, a proliferation in online platforms, such as TradeMe/EBay, Uber, Tinder, and AirBnb, has helped people become accustomed to procuring goods and services in this way. The second is market size: Most people in countries like NZ now have smart phones, providing economies of density in supply/demand (as discussed in last week’s post).

It is not just ride-share: All manner of transport services stand to benefit from turbo-charged IT.

For example, some may think that using different apps for different transport modes is confusing/complicated, and you’d be right. That’s why some clever folks at a Melbourne-based start-up have created Rome2Rio, which is a (gorgeous) ***multi-modal*** transport planner. Rome2Rio searches (mono-modal) transport engines to find the best options for getting you from A to B, whether it be short and long distance by car, plane, boat, or train. For example, below is a possible itinerary to get from my street in Amsterdam to Crete, including details on connections, prices, and schedules.


It is making it easy to travel anywhere, anytime, by any transport mode. On the demand side, by simplifying things people will be able to choose the journey that is best for them, e.g. train from A to B and ride-share from B to C, rather than simply doing what’s easiest, e.g. driving from A to C. On the supply-side, such things are likely to result in increased competition and ultimately see more mileage coming from fewer cars, trains, buses and airplanes.

Of course if/when driverless cars become available then they will seamlessly plug into these kinds of IT platforms.

However, as was discussed in a previous post, if driverless cars are going to be a cost-effective way for everyday people to travel, rather than just a cheaper form of taxi services used for occasional trips, then they will need to carry multiple passengers, just like a bus does currently. And to carry multiple passengers efficiently, then driverless vehicles are subject to the same laws of geometry as public transport: They will tend to travel most frequently along the busiest corridors of demand. Of course they could divert to serve just you, but that’s going to cost you a lot more.

Which brings me nicely onto the next topic …

2. The driverless revolution – How it will turbo-charge PT

Hypothesis #2: The revolution in driverless technology will be applied first and foremost to PT. Cost-efficiencies associated with driverless technology will slash OPEX costs and increase PT’s relative cost-effectiveness compared to all other modes of transport, with the exception of taxis. The latter, however, will tend to complement use of PT in denser areas.

First some background: Drivers are expensive. They account for approximately half the cost of operating a taxi and about 30-40% of the cost of operating PT. So if you remove the need for a driver, then the costs of delivering taxi and mass transit services reduce significantly. In contrast, when you’re talking about private cars the impact of driverless technology on costs is smaller, because the passenger (who was the driver) still has to travel regardless, i.e. the driver’s time is (usually) not “saved”.

Driverless technology may reduce the costs of providing PT services to the point they no longer need OPEX subsidies.

Compared to private vehicles, implementing driverless techology in mass transit systems is relatively easy. PT operates along fixed routes and to fixed schedules. Driverless technology is particularly easy to implement in separated rail systems, which is why so many new metro rail systems are now using it. But it’s not just rail that will benefit from driverless technology, Perth, for example, is currently trialing driverless buses, such as that illustrated below (source).


Meanwhile, slightly further a field one are the runaway success stories of driverless metro systems, which have been constructed in places such as Copenhagen and Vancouver. The success of these systems, which achieve farebox recovery rates around 150%, has encouraged many other cities to follow suit. Milan, for example, has recently opened a driverless metro as illustrated below (source).


Driverless PT, when combined with other initiatives such as congestion charging and accurate parking pricing, raises the prospect that PT services could operate profitably and thereby generate funds to reinvest in more services or infrastructure.

Could driverless PT technologies be deployed in the Auckland context in the near future?

I think the answer is yes. Aside from the rail network perhaps the first candidate for the deployment of driverless technology is the Northern Busway. Don’t get me wrong, I love the Northern Busway – it’s a great service and it was cost-effective to build and operate. The Northern Busway is a key part of Auckland’s successful PT story. And it has scope to grow, with things like double-deckers and more efficient ticketing systems. However, if AC and AT are to achieve their transport objectives, i.e. patronage keeps growing for several decades, then at some point a higher capacity PT technology may be needed.

Driverless rail is one possible solution.

Driverless PT will also put an end to strikes and frequency/span issues that have periodically plagued PT network for decades. I suspect driverless technology may be the circuit-breaker than enables high-quality PT  to operate profitably all-day, every-day. No more Sunday service levels, no more leaving the party early to run for the last bus or train.

At this point some may protest that PT still suffers from the “last leg” problem. Thankfully, this is another area where technology looks set to make out lives better …

3. Sweet mobility – The electric explosion

Hypothesis #3: Rapid developments in electric technologies will not only transform the transport technologies we already have, such as cars, buses, and trains, but it will see the emergence of a range of smaller devices tailored to different needs and peoples. These devices will gradually solve the “last leg” problem and complement high-quality PT services.

Unbeknownst to many, the world is undergoing an explosion in electric transport technologies. This explosion is not, however, confined to cars, even though Tesla garners a lot of publicity.

Ineed, the primary impacts of the electric explosion is on devices that are much smaller in size. For example, here’s the “Folkvänlig”, which for those of you who don’t speak Swedish means “People friendly”. This is an electric bicycle developed and sold by IKEA, which can achieve a top speed of around 30-40km/hr and has a range of 50-100km.

Some of you may be thinking that e-bikes are a niche technology. If so, then you’d be incorrect.

In 2014 global e-bike sales totaled 32 million (source). By way of comparison, in the same year global cars sales were 85 million (source). That’s right: sales of e-bikes alone represented more than one-third of *total* cars sales. Another statistic that worth considering: Global electric car registrations hit only 320,000 in 2015 (source). So on a global basis, annual e-bike sales exceed electric car sales by a factor of approximately 100 to 1. Think of that the next time you see a glib news story about Tesla (NB: I love the concept of electric cars, I just feel that they’re significantly less important than e-bikes).

There’s any number of other wonderful electric mobility devices that are already on the market. Some are even small enough to be picked up and carried with you, thereby complementing the use of PT. We have electric scooters …

And the solo wheel … (“reinventing the wheel”).

Electric technologies also have implications for PT. Buses could soon be electric, while batteries may enable electric rail that is powered by overhead wires to operate “off-grid” for some distances. Such technologies could be applicable in Auckland, for example by enabling us to extend the reach of the EMUs to places like Pukekohe without incurring the costs of electrification.

Conclusion: Reinventing the wheel?

Just to sum up:

  • Hypothesis #1: IT technology will make it easier to use the best transport mode, rather than the one that is easiest.
  • Hypothesis #2: Driverless technology will drive down operating costs to the point where PT is profitable in many cases
  • Hypothesis #3: Electric technologies will improve the transport modes we already have, and see new ones becoming available

In short, emerging transport technologies may mean the future is characterised by far fewer cars and greater uptake of non-car transport modes. I’m not saying this is a given, but it does seem possible – and even likely.

I want to finish by considering a recent report by NZIER. This report evaluated changes in car technology and concluded they were awesome and going to destroy PT, i.e. a different conclusion from what I have reached here. While I disagreed with the authors of the NZIER report on almost every point, I found myself agreeing with some of their recommendations.

That is, given the uncertainty inherent to discussions of future transport technologies perhaps the best thing we can do is ensure that our current policy settings are 1) non-distortionary and 2) durable. Such policy settings are often referred to as “no regrets” policies. So when the NZIER report calls for shifting to a road-user based system, especially one that allowed for time-of-use road pricing, I agree. This would encourage more efficient transport and land use choices today, while also preparing ourselves for the uptake of electric cars in the future, if that comes to pass.

I think this is all kind of cool. There’s lot of exciting transport technologies on the horizon, and regardless of what we do they will result in a more efficient and effective transport system. And even if people disagree on how technologies will impact our future, we can still agree on some changes we should make.

What would you do if you were NZ’s grand transport pumbah? Let’s hear it.

Share this


  1. Point two fascinates me. I’ve long thought it strange that some technophiles think that driverless cars will spell the end of public transport. In doing so they seem to assume that car technology will rocket ahead, yet transit technology will stay dead still. Clearly not the case when cars, trucks, buses, trains and trams all run on the same fundamental systems.

    However I think the key point is that transit operates in an environment with much reduced variables. The driverless metro is a clear example, it runs in one dimension (forwards or backwards) with negligible interference from other actors. Not surprisingly we have had this capability since the 1960s. But even a tram on rails or a bus on a preset route has a much reduced set of variables that an autonomous car which should theoretically be capable of working through live traffic, dodge pedestrians, cyclists, dogs and garbage, and work its way along every motorway, road, side street and driveway in the country.

    The conclusion is clear: Any technology that makes driverless cars efficient and affordable will work better, faster and earlier with public transport. If we can imagine autonomous Google cars on our streets en masse, then isn’t it even easier to imagine autonomous EMUs rolling on the railways, frequently 24/7, costing only the price of electricity to run?

    1. Yeah me too. Driverless technology is great for PT, which is why PT was first to use it in the real-world. And as we see with Perth: Driverless technology is now being applied to buses.

      1. driverless buses, as I’ve said before, I’m not convinced

        as a mass transit technology, sure, as part of a public transport ‘service’, I’m not so sure, how does a driverless bus cope with a passenger with very limited English, who wants to go to an address written in paper on a piece of paper and which they cannot pronounce for a voice recognition system, how does a driverless bus affirm that it is going to the place identified on the destination board (you won’t believe how many times this comes up). how does a driverless bus tell a passneger that they can’t get to their destination from this stop, but they need to go two blocks, turn right and wait for the brown bus (that one’s down the gurgler soon!)

        the driver does more than start, steer and stop, the driver is an interpreter of the PT system for a significant proprtion of the punters

        then there was the bit I read recently, that Mercedes Benz are replacing robots in their factories with people, because Mercedes cars are becoming much more individualised and people deal with complexity much better than robots, now that I’m back behind the wheel of a bus again, I have time to muse on these things, sure aspects of driving can be reduced to an algorithm, but bus drivers manage a great deal more than a driverless car

        how does a driverless bus check for lost property? securely collect it and report it to an operations centre? nope, not convinced at all

        1. I have fairly strong views on this: You don’t design a PT system around the lowest common denominator (in terms of passenger needs). Instead, you design it to be as efficient as possible and if that doesn’t suit everybody then there’s other options available. If someone can’t use a bus, then they can’t use a bus.

          The other thing to consider is that in 10-20 years time, which is when I imagine this technology will be ubiquitous, then today’s 40-50 year olds will be 50-60 years old. I.e. the older people will be used to smart phones. That’s what tells you where the bus is going, not some poxy display board :).

          1. And a lot of these issues can be addressed by better information, for instance maps and timetables at all stops (as at stations), and on-board identification of the next stop (as on trains).

            In many parts of the world bus drivers do indeed just drive (sometimes even in a separate cab – standard practice on modern trams), so I’m not sure why NZ needs to be different.

            But I’m not sure how a driverless bus knows the difference between a standing pedestrian and a waiting passenger, or knows what do to when a bus stop is blocked or it cam’t get to the kerb. Clearly a higher standard of bus infrastructure will be needed.

          2. Easy: By the time buses are driverless, then you’ll book them on your phone while waiting at the bus stop. The next one to come along will stop if it has capacity.

          3. I think it’s dangerous to presume that every single bus user will have a working example of the right technology on their person whenever it’s needed. There will always be people for whom mainstream technology is difficult (or just have a flat battery), and to write them off is to my mind a big mistake.

            And what if I summon the bus in this way but I’m not there when it arrives? (eg I’ve got a lift, or changed my mind). Does the bus stop and wait? Does it drIve on if it doesn’t detect anyone, though I may just have moved to sit down, stand in the shade or out of the rain?

            Lots of imponderables – this needs a lot more thinking through.,

          4. part of the issue is that the term bus covers a broad range of service types from true mass transit (think Transmillenio) to small scale community level services, (often carried out by 42 seaters because of the practicalities of service provision)

            at the upper end driverless buses may be applicable, at the lower, I still think that the non-driving personal level services and assistance (helping frail and elderly with shopping, mums with buggies etc.) provided by the driver will remain a given

            and left property (relatively valuable stuff like smartphones and tablets, without which you can’t find the next bus in your scenario) remains a daily day issue on the Northern Express, as does “does this bus go to Sunnynook?”

          5. Hi Mike, I think you’re possibly at risk about jumping to conclusions about what I think.

            More specifically, I’m not “writing people off” if they don’t have a smart phone. Instead, I’m thinking more broadly about what a future PT system may look like, rather than getting trapped by the “what we do now determines what we need to do in the future.” Keep in mind that we’re talking about 10+ years here, i.e. a long time in technology land.

            In 10+ years’ time, for example, PT services will probably be cash-free. Instead, passengers will be paying via HOP or some other smart device, like a phone. In this context, I suspect it would be quite easy to design a system which had a HOP card/device reader at every stop where people could swipe their HOP card and let the system know which service they’re waiting to board. In mixed corridors the system could incorporate functionality so that people could enter the specific route number they were waiting for.

            In fact, you could argue that such a system might be useful now so as to avoid the situation where people have to wave down the bus, which I’ve always thought is a bit of a pain and something that would go by the way-side as patronage grew and technology improved. This is not only challenging for drivers when there’s lots of people on the footpath, but also challenging for many vision impaired people, who obviously can’t see when the bus is coming.

            And such a system would be a relatively small price (say 2,000 stops times $2,000 per stop = $4 million) to pay compared to the cost of having drivers on our buses.

  2. Brilliant post, Stu! Really nice job linking together a lot of important trends.

    I think the autonomous vehicle discussion has got it wrong by focusing on cars rather than trains, buses, and trucks. It’s reasonable to expect uptake to be fastest among fleet operators, especially bus fleet operators. And I can’t see why cutting the cost of providing PT services by 20-30% would be bad news for urban PT services. (Or even intercity bus services, which are already very cheap.)

    Another random thought: Bla Bla Car and Rome2Rio are examples of the importance of agglomeration economies at multiple geographical scales. They probably work best in places where:
    * Cities are relatively compact, allowing people to share rides point to point more easily
    * Cities are relatively abundant and close together
    * Transport networks are dense but also relatively well-used.

    Three conditions that NZ doesn’t really meet…

    1. Thanks Peter – as you know from our time spent skiing and drinking beer in Switzerland this post has been brewing for a while 🙂

      Yes I think you’re right wrt agglomeration economies contributing to the usefulness of some IT technologies.

      However I can see them being transferred to the NZ context over time, especially in the upper North Island. I’m sure there’s quite a bit of actual and suppressed demand for travel between Whangarei, Auckland, Hamilton, and Tauranga for example. Also NZ has a lot of tourists tootling around in rental cars who may be willing to share their car to offset some of the costs.

      Basically, I think things like Bla bla car will start in places like Europe where they’ll grow their user base and refine their platform. Over time however I’d expect it to be transferred to NZ and not without some success. I would also expect Bla Bla Car to increasingly compete for short-distance journeys, once they’ve nailed their long-distance offering. So my general message is “watch this space”, as I think it will quickly become useful even in NZ context.

  3. I find the “romantic partners” graph really interesting as well. It surprises me that it shows such a high proportion of people meeting their partners through university back in the 30s, when relatively few people went to uni and the gender ratio was, I imagine, more skewed towards males rather than females. I assume the surveys were done on university students rather than the general population, although they’re still fascinating.

    1. Yes I’d focus on the trends rather than the levels, because the latter is more susceptible to sample bias. Not sure what the sample was in this case …

      Do note however that the graph is only up to 2009. There’s been a lot of changes in internet dating 6 years, e.g. Tinder.

      1. I think it would be pushing the definition of the term “romantic partner” to apply it to the people met on Tinder…

        1. you’d be wrong. Many people have met partners on Tinder.

          And how do you define “romantic partner” anyway? In my experience Tinder can be fairly bloody romantic.

          1. I think your ignorant arrogance is (again) getting the better of you.

            “Hooking-up” is one possible step on the road to a romantic relationship. Some people who start by hooking-up go on to have enduring relationships with the people they met online.

            Here’s what a researcher had to say on the topic:

            Melbourne University’s Dr Lauren Roswarne, who is currently writing a book called Intimacy on the Internet, says that every method of finding a partner has a series of unknowns. “While we have long feared — and stigmatised — people who met on the Internet, such people are no more creepy or dangerous than people who met in bars; the anonymity of the Internet, however, along with technology more generally has long scared us,” she told Mamamia. “Technology might be designed for one function but in practice get used completely differently: hook-up apps are a good example; they were designed for setting up location-based hook-ups but have come to be used by some people to find a partner.”

            P.s. My fiancee and I met on Tinder.

      2. It’s also extremely helpful to think of that tinder graph for a number of examples over time.Think about how people used to book hotels/motor lodges, get directions, book a taxi. A few years ago I cold probably whip out an AA travel lodging guide, road atlas or recite the local taxi’s phone number (to the tune of a radio jingle). Information technology fundamentally challenges how people access information and the powers that used to monopolize that information.

        Hypothesis #1 for me is the biggest game changer, Waze and google maps push me traffic conditions and road closures, moovit and citymapper push me notifications for PT rerouting and special events. All of this access to information enables me to make the best travel choices specific to my needs without strenuous planning. Complementary services like Uber or Car2Go provide redundancy when it starts to rain, or I’ve missed the last bus, providing assurance and reliability regardless of my choice of mode.

        Great articles Stu.

        1. yes I agree – provided that the provision of information is not monopolized by a new intermediary like Google :).

          I do think however that there will be a healthy degree of competitive tension in the IT space. While apps like AirBnB and Uber definitely have an advantage from their scale, the barriers to entry really aren’t that high – and certainly lower than they would be to starting, for example, a new taxi company and/or hotel chain.

      3. Hard to see how there could be a small surge of people meeting online back in 1984. Strange also that the war years didnt have any real difference.

        1. Dance halls were very popular after the war and well into the mid 60’s, how many dates resulted in marriage I don’t know, but it was a sure fire way of meeting the opposite sex.

  4. An excellent article, my interest is in electric bikes they need very little in the way of infrastructure, they can run on tracks or roads with far lower cost to maintain, my favourite bike is my wife’s tadpole trike, it has the driving wheel at the back, with 300 watts of power it goes anywhere a car could go.

    I wonder about this driverless technology whether it’s not another nail in the coffin of the working man with more wealth congregated in the hands of the few, but having said that from the point of view of the cyclist it could be wonderful as there would be no need for parking at the side of roads as the vehicles would be in constant use with no privately owned cars on the roads, would make for a great future.

    1. Slightly OT, but what sort of performance do you get out of 300W? I have been thinking about building an E-bike for my wife for a while but am not sure of what power level to get. I know 300W is the maximum legal power but lets leave that aside for a second. I live in Birkenhead where the hills be steep, but I otherwise wouldnt need it to go too fast – 25kph with some pedal assist would be all good.

      1. Go down to Browns Bay and test ride an e-bike with the folks at Bute Bike.
        There are plenty of steep hills there to test on.
        To answer your question, 300w is good. You’ll zip up hills with greatly reduced effort.

      2. Yes scott is right you need to try a few, be have a 250w E bike and the 300w recumbent trike, I would go for 300w if you have hills they have variable power settings from 1 to 5, giving a speed range without assist of around 10kph to 30kph on a 300w, I have a few videos of ours on youtube at bandjsellars.

        One of the problems we found was bikes without crossbars are not as stable when you come to stop, the extra weight on the back from the battery can throw you of balance, it may not be a problem if you are young but we sold a couple of folding E bikes for that reason.

  5. Cracking post Stu. Excellent summary.
    The reason for the hoopla around driverless and electric tech being so focussed on cars is of course two-fold:

    1. The huge marketing savvy automotive industry ‘talking their book’, and
    2. Path dependency in our transport institutions.

    The former needs no further exploration, but the later is a real problem; things have been improving with this recently (especially around cycling because of top-down policy change) but still all our major public and private (eg NZIER) transport agencies are essentially focussed on what they are habituated to see as their primary ‘clients’: the private car user and the truck companies; this has become almost the entirety of what the word transport is considered to refer to. This habit results from the monotonal road building age the west is only now emerging from; for almost everyone’s entire careers have occurred in this bubble. It is also a bias resulting from an unsophisticated application of the ‘user pays’ ideology, an often wilful denial of the externalities of road transport (certainly wilful by trucking lobbyists). This has been these institutions’ path in living memory; and to a significant degree they have changed shape to meet it. They are on this path and they are largely dependent to it. It will take effort to reform this institutional distortion.

    So of course driverless and electric technology is feverishly researched by these institutions for cars. They are, still, largely blind to other modes. This is a clear sign of group-think, and path dependency.

    For example; is there a well resourced workstream anywhere; MoT, AT, NZTA, looking hard into driverless trains for AKL and WGTN? For electric buses?

    Why not? Isn’t it a clear means to radically improving value for money? Subsidy reduction or removal. One of the key criteria in founding documents, including ATAP.

    1. PR: yes, there is work taking place on electric buses in Wellington, planning for their replacement by diesels next year. Duh….

      As for driverless trains, they exist only on discrete single-purpose railways with homegenous rolling stock and no level crossings, none of which apply to NZ (except for two short urban branch lines). There’s an awful lot of work to be done before the world is ready for driverless passenger trains sharing tracks with freight trains and crossings with road users (of all types). .

      1. Mike Stu’s argument above still holds; if road vehicles can be autonomous then it stands to reason that trains can be sooner, even on mixed lines with level crossings, as this is a much simpler environment with fewer variables than any public road.

        My, fully non-expert, understanding is that ETCS-3 is essentially what is required, from Wiki:

        “With Level 3, ETCS goes beyond pure train protection functionality with the implementation of full radio-based train spacing. Fixed train detection devices (GFM) are no longer required. As with Level 2, trains find their position themselves by means of positioning beacons and via sensors (axle transducers, accelerometer and radar) and must also be capable of determining train integrity on board to the very highest degree of reliability. By transmitting the positioning signal to the radio block centre it is always possible to determine which point on the route the train has safely cleared. The following train can already be granted another movement authority up to this point. The route is thus no longer cleared in fixed track sections. In this respect Level 3 departs from classic operation with fixed intervals: given sufficiently short positioning intervals, continuous line-clear authorisation is achieved and train headways come close to the principle of operation with absolute braking distance spacing (“moving block”). Level 3 is currently under development. Solutions for reliable train integrity supervision are highly complex and are hardly suitable for transfer to older models of freight rolling stock. Some kind of End-of-train device is needed.”

        Moving Block systems are used in driverless trains in London [DLR], Singapore, and Vancouver.

        Is there work at MoT about the opportunities to apply Moving Block systems to our passenger and freight systems in urban areas, it would increase efficiency even if it didn’t mean full autonomy? Costs, Benefits….?

        1. I think many people in the know would challenge your assertion that mixed-traffic grade-level rail is a simpler (let alone much simpler) than road. They’re very different, with rail being much more systematic and risk-averse, and in those respects being more complicated. And retrofitting rolling stock and replacing/upgrading signalling systems is much more complex and expensive than just adding new vehicles to the road.

          I agree fully with Stu in principle, but experience shows that the practice is not necessarily the same, with the only driverless rail examples on the horizon being fully grade-separated homogeneous metros, like the examples you quote (and which are by no means universal – eg it will be many years before the London Underground has *any* driverless trains).

          For instance:

          – there are driverless cars already in opertation on public roads; nothing similar is happening with rail (implementing ETCS1 and 2 is proving problematic enough);

          – driverless technology is being developed largely in the US, but their mainline freight railroads are still debating whether/how to remove one of the current two people from the cab. Diverless is nowhere near the agenda, even with the mandated PTC (their ETCS equivalent).

          Organisations have become a cropper over moving-block technology (one of the factors causing the UK’s Railtrack to go bust), so I’d expect to happen only where it’s really required and where it the environment supports it. Neither applies in NZ at the moment, but I hope (and expect) that MoT/NZTA/AT/GWRC/Transdev are all aware of the possibilities and prepared to be a reasonably fast follower.

          1. I guess the question is what is the reason for the different rates of technology between modes? We know, for example, that the reasons that London is still so driverful is much more political than technical. It is a fully grade separate passenger railway, on most lines anyway, but unions are powerful its current operation.

            Additionally driver cost on the huge US freight trains is probably much less of a proportion of cost that other fixed factors, and they are putting huge effort into automated control largely to reduce fuel use and optimise efficiency and timeliness.

            Many huge coal trains in Australia are automated, this isn’t difficult, in those isolated environments [at the cost of the odd Kangaroo, presumably].

          2. Sorry to argue, but we don’t know that the reason London still has drivers is more political than technical, because that’s a politically-spread myth. Installing signalling that is consistent with automatic operation has been a very fraught process, still very far from complete, and very expensive with hundreds of millions wasted in false starts. There is also the safety issue of potentially having no staff member on the train – and if your still having to employ on-board staff, the financial justification starts to slip away….

            Tube drivers are certainly active and we’ll organised, but that’s merely one part of a much bigger and more complicated jigsaw.

          3. And American railroads are putting huge efforts into automatic operation (but still under the control of a locomotive engineer) because Congress will close them down if they don’t. The financial return from PTC is in fact pitiful, a classic example of making the safest mode spend money to make itself safer but less competitive because it’s easy, while less safe modes carry on in their own sweet way. That would never happen here, would it?

          4. Hi mike thanks for the useful comments. I agree driverless operations on exisiting rail networks is tricky. My point however is that in the next 20 years can we not progressively upgrade parts of our rail network so as to potentially enable such operations? E.g. gradually address level crossings and seperate freight and passenger rail. Either way northern busway could be a standalone route..

          5. Thanks, Stu – agreed.

            I think the basic issue is that roads being an open system things like driverless cars can be introduced incrementally, while the closed nature of the railway means that the whole system has to be upgraded first. Add to this the need to interact with open systems (eg at level crossings) and with other parts pf the closed system (eg AT’s operation relative to KiwiRail’s), and things start to get just a bit harder. So, taking early steps like level crossing closures would indeed be very helpful.

          6. yes and that’s an excellent point; so thanks for the contribution. And I think we’re rapidly heading towards consensus.

            My goal here was to simply highlight that this technology presents significant long-run advantages, e.g. AT are currently spending $150 million p.a. operating passenger rail services in Auckland. If we assume 20% of this is driver-related costs then this equates to say $30 million p.a. saving. Over 20 years this would have a discounted value of say $400 million.

            In the context of Auckland’s rail network, I wanted to highlight that *if* we’re already investing in things like the CRL, upgraded signalling systems, the third/fourth mains, and separating rail level crossings, *then* we may want to consider spending a bit more here and there so as to potentially enable/expedite driverless operation at some time in the future. I wouldn’t make driverless operations the be-all-and-end-all of what we do now, but I would consider it in our strategic planning because it seems to me that there’s potentially some things that we could do now which would make it much easier later on.

            You’re absolutely right that we should not misrepresent the challenges involved in achieving a driverless passenger rail system. Hopefully by the time Auckland considers this question the conversion to driverless technology will have become more common so we can learn from others experiences.

    2. What would I do if I were NZ’s Great Transport Pumbah [Poo-bah?]? Consistent treatment of all transport infrastructure would be top of my list, including capital expenditure and pricing. Free goods, such as road use, in a competitive situation create distortions with significant negative externalities, and and excellent use of new (and existing) technology would be to address this, eg with time-and-place-specific road pricing, for a start.

      1. have to agree with mike here, Ive worked in the rail industry for nearly 20 years and the side effect of safety gone mad is increased cost to operate with the flow on being modal shift towards road freight in many cases. shunting sidings was deemed dangerous so trucks do the work instead basically.

      2. Glad you kind of asked. Pumbah is a personal hybrid term that mixes the more common “poo-bah” with “pumba” (who is an endearing warthog that featured in the lion king). I like to mix my metaphors because i find it funny to think of a warthog in charge of nzs transport system. Wallowing in mud and concrete etc.

  6. Great post Stu. Have you investigated Mobility as a Service technologies?

    MaaS makes moving people from A to B into a service by unifying active, private, public and shared modes of transport into a single gateway for consumers. MaaS trials indicate that this can lead to a disruptive reduction in car ownership with driverless vehicles further disrupting the ownership model. Finland is leading the way on MaaS and launching a new service this year. In the future the form factor of vehicles will change to single commuter vehicles as an efficient means to move individuals, people-movers will replace buses and trains will remain for mass transit of people being main centres.

    The below paper is analysed from MaaS delivered as a profit making venture with transport network efficiency being the by-product .

    If I was the Minister of Transport I would be investing in delivering MaaS with transport efficiency as the main driver. This technology could get Auckland traffic moving in the near future and massively reduce the debt-funding required to build roads.

    1. Thanks toa and glad you enjoyed it. I think its interesting that the future i sketch here can appeal to everyone: all transport users will be better off no matter what modes u use. I honestly believe that it seems likely all of the transport modes we use now will continue to be used, albeit with technological enhancements and changes in emphasis depending on their comparative advantage.

      No haven’t heard of that MaaS concept but it seems similar to others, such as “complete mobility”. But the idea is, i think, the same: people decide when and where they want to travel and how they get there is constrained by preferences and means rather than mode.

  7. What’s interesting as well is that the *dimensions we focus on* determine what we regard as transformational change

    For example, if we looked at (for example) speed, reliability, and time between refuellings, there hasn’t been a dramatic change in the automotive industry for 50 or so years

    But when we look at other aspects e.g. control mechanisms, fuel source, we see major change

    Will these three trends be as revolutionary as say the invention of the railway – or will they just step change things (as with the move to diesel and electric locomotives)

  8. And, if I were NZ’s grand transport pumbah
    1. I’d set a range of targets about commute times, sustainability, harm (deaths and injuries), and reliability
    2. I’d institute fully flexible output-based costing
    3. I’d then let my sub-pumbahs loose, and sack the ones that didn’t make their targets
    4. Rinse, repeat, until clear targets and flexible funding leads to the right mode for the right place and for the right target

    1. Fine, as long as you set appropriate priorities on your quite disparate targets. Unlike today, where for example, rail-safety has to clear a much higher bar than road, as does rail funding, and mitigation of rail-externalities.

      And this is where things gets difficult in the ideal-world of objective targets, because human irrationality supports these glaring disparities and sees nothing wrong with them. So much of what we do is skewed by subjectivity and this includes determining transport policy.

      Steven Joyce loves roads, and so finding himself as NZ’s grand transport Pumbah he bulldozed through the economically-inefficient Roads of National Significance. You can wax on all you like about “right mode, right place and right targets”, but human reality is that “right” is very subjective and people will do what they want (or what is right by them) if they can get away with it. And the only way to counter this is by people-power, lobbying, activism and effort. Welcome to the messy world of policy-making.

      1. This is copied from a report by the NZTA in 2013.

        (The cost of maintenance, operation and renewal of roads in New Zealand has
        increased at a rate well above general inflation for the last ten years.) it didn’t say by how much though.

        They say rust never sleeps, neither does wear and tear to our roads, maintaining roads is not something I think most young people will want to fund when they have little prospects of owning their own home or wish to own a car, especially if the cycling culture grows, I certainly don’t.

        No trucks should travel more than to a rail depot, electrified rail has a history of over 100 years and has a proven record, I think it’s time we caught up, but there again I’m old and remember a better time and it’s not looking back through rose colored glasses as some may think.

  9. The electric device for the last leg is very suitable for new Zealand.

    However our adoption rate is very low compare to china.
    Part of it due to inflated cost here.

    In China, a electric bike is only about $500-$800 NZD. However in NZ, they are selling in $2000+ NZD.
    Unless there are more demand, there won’t be enough competitors to force the local retailer to drop the price.

    The second issue, is the legal uncertainty

    For example can you ride a slow hover board on side walk? Can you take it to train and bus? Do you need helmet? What is the maximum speed?
    Is the same rule apply to solowheel? How about e-scooter? What about segway like device?
    Early adopters fear their investment to such devices suddenly become illegal to use.

Leave a Reply