Here’s a new study from MIT that says ride sharing and pooling algorithms could theoretically reduce Manhattan rush hour traffic drastically.

On-demand high-capacity ride-sharing via dynamic trip-vehicle assignment

Ride-sharing services are transforming urban mobility by providing timely and convenient transportation to anybody, anywhere, and anytime. These services present enormous potential for positive societal impacts with respect to pollution, energy consumption, congestion, etc. Current mathematical models, however, do not fully address the potential of ride-sharing. Recently, a large-scale study highlighted some of the benefits of car pooling but was limited to static routes with two riders per vehicle (optimally) or three (with heuristics). We present a more general mathematical model for real-time high-capacity ride-sharing that (i) scales to large numbers of passengers and trips and (ii) dynamically generates optimal routes with respect to online demand and vehicle locations. The algorithm starts from a greedy assignment and improves it through a constrained optimization, quickly returning solutions of good quality and converging to the optimal assignment over time. We quantify experimentally the tradeoff between fleet size, capacity, waiting time, travel delay, and operational costs for low- to medium-capacity vehicles, such as taxis and van shuttles. The algorithm is validated with ∼3 million rides extracted from the New York City taxicab public dataset. Our experimental study considers ride-sharing with rider capacity of up to 10 simultaneous passengers per vehicle. The algorithm applies to fleets of autonomous vehicles and also incorporates rebalancing of idling vehicles to areas of high demand. This framework is general and can be used for many real-time multivehicle, multitask assignment problems.

There are plenty of criticisms of this type of study. The major one is that if you make a particular transportation option faster and/or cheaper, economics dictates that people will automatically switch to it from other options over time, eventually making it less fast and/or less cheap until the various modes are balanced again. The study above (based on my quick skim of the abstract) probably took data from one or a few Manhattan rush hours and asked how it could be rerouted in the most efficient possible way. I don’t fault them for doing the study, which is really interesting. The economics and human behavioral feedback loops that happen over longer periods of time just need to be studied too before policy decisions are made based on results like these.

I don’t necessarily want UberPool to be the answer to all our infrastructure problems. I love the idea of subway and above-ground rail and bus rapid transit as much as the next person. But as the opening of the most recent segment of New York subway recently showed us, these projects are taking decades to build in the U.S. and costing enormous amounts of money. Europe and Asia are doing much better than us, so maybe we could learn some lessons from them, but our recent political challenges shed some doubt on the idea that we can improve any time soon. (Europe generally manages to do somewhat better with high-wage union labor, while some Asian countries build extremely cost-effectively by issuing temporary work visas to low-wage labor from developing countries. There are political and moral issues on both ends of this spectrum, obviously, but the point is the U.S. doesn’t do either approach well. Much like our health care system, we spend 2 or 3 or 5 times more than everyone else and get worse results.)

If the criticism of the study I mentioned above is that demand projections made before the new infrastructure options or technologies are in place are not going to be accurate, that criticism certainly applies to a subway system that takes decades to build. The entire population, land use, and employment pattern of the area served could change in that time, not to mention that whatever technology is chosen is almost guaranteed to be obsolete the day operation begins. With the ride-sharing algorithms, even if the projections are wrong at first at least you have a system that should be easy to adapt and tweak over time. I don’t see why public bus systems and bus rapid transit can’t be integrated into a system like this. And if people want a vehicle to themselves for some trips sometimes, the algorithms and pricing schemes should be able to accommodate that. You could even imagine an algorithm managing passenger vehicles, freight and delivery vehicles in urban areas so they are less in conflict with other at various times of day and night. The algorithms could be run by government or non-profit entities if we are really afraid of private control, or private algorithms and entities could be forced to communicate and coordinate with one another.