Friday, December 9, 2016

Public transport and driverless cars: How they will play.

By Andrew Atkin

Many people believe that driverless cars are decades away. They have not followed the technology closely, or thought seriously about its deployability.

My example is the twizy car. It weights about a quarter of a modern sedan. Now imagine having this car in a car-sharing scheme whereby it can drive itself to the next customer, though you drive it when you're in it.

How far away is this car-sharing revolution, if that is all you need to provide for a major market impact? It should be with us in a year or two, in principle. The technology can already do this.

There are no serous safety concerns with a system operating like this when the cars are so light, and no one is in them when in driverless mode.

Now take the picture of little twizy cars everywhere, costing a fraction of conventional cars to use, and using about 20% of the energy of buses, trains or cars. Where does this leave public transport as we know it?

Well for a start, a driverless twizy system is perfect for solving the last-mile problem of buses and trains. In particular it's good for supporting buses, as the supply for the demand is more spread out.

But driverless twizy's will of course cover much more than the last mile. They will replace conventional public transport wherever they realistically can. Though where driverless technology can't go, public transport will survive.

Public transport:

The only places where public transport can survive as a competitor to driverless cars (car sharing) will be in areas where the transport demand is high enough so that congestion becomes an issue.

Assuming that congestion-charging is employed to control for congestion, certain roads will be costly at times to use. This is where buses step in. People can avoid the expensive road toll by using buses as a point-A to point-B bridge. This is somewhat easily done as passengers can switch from the driverless twizy to a bus, at both trip ends, and maybe in as little as one or two minutes.

So this is how it would work...

Buses will be dead except for where they can provide capacity bridges for high-demand roads.

But to the end of providing capacity, buses can be very efficient in this context because:

1. They're operating on congestion-free roads.

2. Have high average passenger loadings.

3. Have rapid trip times due to minimal stop-and-go (only 2 or 3 bus stops at each end of the major trip).

Certainly you should not require subsidies to support public transport in a driverless reality, if they stick to their sensible role.

Demand for collective transport:

However, buses and trains will have to compete with other forms of collectivised travel.

How about shuttle buses? And what about car-pooling specifically as a driverless service?

A driverless car can be built as, say, a 6-seater, with retractable partitions for privacy (ref. yellow lines on image). That will be cheaper and more efficient than a bus, be great for low-cost long distance travel, and provide a more responsive service.

Driverless car-pooling would probably make more sense in most applications than a bus, to the end of providing capacity relief as required. Many small vehicles are always going to be better than a few large ones, insofar as capacity demands allow for it.

End game:

It's very difficult for me to imagine how buses and trains can survive long term.

There will surely be demand for collective travel for where it makes sense as a cost solution, but buses and trains will probably prove to be an intermediate solution that will only survive until driverless solutions, plus some inevitable infrastructural upgrades, finally render them redundant.

It can't be under-stressed as to how close this movement is. Again, all that little twizy car needs to do is be allowed to transport itself to the next customer, and it's the beginning of the end for transport as we know it. The transition to driverless-based car-sharing will be rapid. The cost reductions and convenience of this kind of car-sharing are great, and it can absorb most travel demand pretty much immediately.

So to conclude, conventional public transport, as a solution, will serve as a capacity bridge and little more than that. And that somewhat minor demand will most likely prove very temporary as well.

1 comment:

  1. Think of this: If you build a road that accommodates car-pooling 6-seater driverless cars only, and not heavy truck convoys, then you create a road (namely, under or over passes) that will cost maybe a third of a conventional road, yet delivers about 4x the peak capacity. That dramatically reduces the real costs for infrastructure.

    This is why we can comfortably presume that the demand for buses and trains will collapse, long term, in a driverless reality. People will surely demand the alternatives once costs are rationalised.

    -Andrew Atkin