When Cars Fly
Why I don’t laugh at Flying Cars: the Economics and Physics Make Sense
Yes, they look funny. We can all agree on that.
But the physics and economics behind flying taxis actually make sense.
First, why do they look so funny?!
Noise is the reason why of the 23 helipads in San Francisco, only one is operational. Noise level is the fifth power of rotator speed, the result of the blade ends hitting air. Helicopters need that rotator speed in order to generate the lift it needs. However, if we split one rotor into many, then each rotor only needs to generate 1/n amount of lift, lowering the rotator speed and size of the blades. Now you know why they are shaped the way are.
But the physics of lift isn’t new, and the 1/n number of rotors isn’t hard to figure out. Why now?
Years ago in a multiple-rotor setup, if one rotor failed, it would immediately shift the lift balance, causing flips at best, catastrophic failures at worst. Software and processing speed have caught up enough to load rebalance immediately, in real time.
The current models are almost all electric. With the speed of advancements in battery technology, we should get to 60–100 mile in one charge within the next five years.
Lastly, who’s going to use these things?!
I’ve sat in traffic in Jakarta … even 60min in an air conditioned van was intolerable. Despite being in suits, all the passengers in that van decided to get out and walk the rest of the 20 blocks, even in 100F degree heat.
I’ve sat in traffic on the highway in Beijing during Golden Week, 45min extra on the ride from the airport … even at midnight!
I’ve sat in traffic in Rio de Janeiro, on my way to the airport, even at 5.30am! I personally know friends who use helicopters to commute to work in Rio.
I live in the Bay Area, where the 30min drive between SF and Palo Alto, can easily turn to 2 hours during rush hour.
So I assure you, there’s demand. The real question isn’t demand, but economics.
Currently, it costs around $10/mi helicopter passenger mile.
We can cut around a third of that cost off if electric.
If we can improve utilization, and have a high load factor, that cuts down the cost to $1.50 per mile, which is around the current cost of UberX. By the way, an extreme example of utilization is space travel: non-reusable rockets is an extreme example of low utilization and load factor.
Airplanes are currently mostly built and assembled by hand. If we can scale manufacture planes like cars, we can cut down the cost further to $0.50 per mile, which is below today’s car ownership (Uber’s Dara noted in 2018 that the current cost of car ownership is around $0.7–1.00 per mile).
The explanation above is why I believe flying taxis may be an inevitable eventuality. But I also see bottlenecks that need to be solved before they become viable transportation alternatives.
(1) Security: I lived through 9/11, so security is top of mind. Depending on the speeds, how much passenger screening and security should we have? There are currently millimeter wave technology that can detect weapons and carries without the customers even stopping.
(2) Autonomous: If you think autonomous driving is hard, autonomous aviation may be even harder. It’s in a 3D space, with wind speeds, fog, and catastrophic failures mean falling objects from the sky.
(3) Regulation: Currently the FAA is requiring thousands of rural flying miles before they will let any players fly urban miles. The first urban miles are also likely to come from delivery drones, rather than passengers.
(4) Utilization: This was the crux of getting the economics to work. Chariot was a commuter bus startup that Ford paid $65 million for in 2016, only to shut it down 2.5 years later. The flying taxi industry would be wise to learn from Chariot’s path: growth steadily, make the unit economics of each plane work, even if it means growing slower.