5 Oct 2014

Public Transport in the Age of Smartphones

Public transport systems like metros and buses were designed in the industrial age: the transport agency would do a study, and lay a new metro line or bus route, and people would take the train or bus, perhaps with a route map as an aid.

Google Maps (disclosure: I work for Google, but on nothing Maps-related) dramatically simplified this for passengers. I can go to any city in any country of the world, and Google Maps would tell me the best route to reach where I want to go, across multiple transport systems like metros and buses. Some transit agencies have their own official apps, which can give more information than Google Maps. But these apps are at best an add-on to an already existing transport system.

But now, in 2014, when many people have smartphones, how can we build a better system by taking smartphones into account as an integral part of the transport system?

For example, we could intelligently group people into buses based on their destinations, so that each bus can make fewer stops along the way, resulting in a faster commute for everyone. When you reach the bus stop, you’d tell the transport app where you want to go, and it might tell you to take a particular bus at a particular time, grouping people based on the destination.

As a simplified example, assume there’s only one bus route that goes through the bus stop you’re waiting at, so you could take any bus that shows up, but the app might tell you skip the next bus and take the one after that, if that’s most likely going to be faster for you.

In addition to a faster commute for everyone, this system makes for a more pleasant ride due to less acceleration and deceleration, less energy consumed, and a greater capacity of the overall bus system [1].

At its peak, this can become a point-to-point system, with no intermediate stops. If you’re standing in a bus terminus during rush hour, and want to go to M.G.Road, and there are enough people who want to go there as well, all those people could be given their own bus. You’d punch in your destination in your transport app, and it would tell you to go to bay 12. Immediately, the bus there has its sign lit up, and programmed via the network to show “M.G.Road”, rather than a route number, because the bus is not on a predefined route. Everyone who said they wanted to go to M.G.Road would be asked to go there, and the bus would leave.

In addition to skipping stops, this kind of system can also have buses stop at a point in between two formal bus stops, if enough people want to get off there, say at an office building in the morning rush hour. This will save time.

Similarly, a bus could skip a stop if no one is going to get off there. Ideally, your phone would signal to the bus which stop you will get off at. No more anxiously looking at Google Maps trying to figure out when you should press the button to stop the bus.

In addition to stopping at different places along a predefined route, this kind of demand-responsive system could reduce the need for transfers, by assigning a bus that connects stops on two different routes if there’s sufficient demand at a certain time.

Another example of a smart transport system is one that detects a bus where all the seats are taken, and there’s at least one person forced to stand, and immediately dispatches another bus from the terminus. This is better than the alternative, industrial approach, which is going by a predefined frequency. A predefined frequency is problematic because the planners can’t perfectly foresee demand months or years in advance, so buses often run empty or force passengers to stand. Instead, dispatch another bus on demand.

Metros can also be configured to have more or fewer coaches, depending on the demand. At off-peak hours, metros should run with only one coach. That is, it’s faster for passengers to have a one-coach train arrive every five minutes than it is to have a three-coach train arrive every 15 minutes. So, at off-peak hours, metros would be configured with only one coach. And the frequency would again be determined dynamically based on demand.

As the day begins and traffic increases, the one-coach trains would keep arriving more and more often. Eventually, they’d reach their maximum frequency, at which point the system would automatically transition to two-coach trains. And when those are not enough, three-coach trains, and so on.

Another idea would be: can we have smaller vehicles than buses or metros, like vans that seat less than 20 people? That way, you’d have a greater frequency of service. What size of vehicle would let you achieve your desired standard of service at the least cost? Alternatively, what size of vehicle would let you achieve the best service at a reasonable cost?

For example, Bridj makes a trip in 20 minutes what the train takes an hour. It does cost $6 rather than $2.5 for the train, but many people would consider that worthwhile to save 40 minutes. There are also other van startups like Chariot. Some of these startups have been called the Uber for buses.

Then there are plain old shared taxis. In India, where the salary of a driver is low, there’s no reason they can’t compete effectively with other modes of transport. I can imagine something like UberPool, but where an Indica shows up rather than a higher-end car. It would seat three people, in addition to the driver. This means that it would make only two more stops to pick others up. With bigger vehicles, you need to make more stops to pick people up, which slows down your commute.

Another approach is to buy a 7-seater van for not much more than an Indica, more precisely, for around ₹4.3 lac on-road, so carrying 6 passengers at once will be cheaper than carrying only 3. After all, neither the car nor the driver costs more, but it carries more people, so the per-passenger cost comes down.

Public transport systems need to move past the industrial mindset. There are many ideas to improve public transportation when you design systems taking into account that people have smartphones, and design for it. This will result in commutes that are faster, cheaper, more comfortable and green. And, once again, the future is being built in the US.

[1] If the capacity of a bus is 50 people, but the bus is able to serve those 50 people quicker, then it’s ready to serve another set of passengers, resulting in an increase in the overall capacity of the bus system. Which in turn means quicker service, and less crowding and standing during rush hour. And less investment needed for the same level of service.

No comments:

Post a Comment