The Last Mile of City Drone Deliveries

By Tom Farrier, Former Director of Safety, Air Transport Association. Originally published on Quora.

Drone delivery proposals are just one strategy aimed at addressing a larger business challenge generally referred to as the “last mile” problem. Why is the last mile so inefficient? offers a brief, non-drone-related explanation of this issue.

In a related article (9 trends in last-mile delivery), drone delivery is mentioned as a potential game-changer. However, it acknowledges one current challenge to all such possibilities: “The robot needs human accompaniment in case it has a problem.” To me, that is a major issue that the sexy pitches for drone deliveries really tend to gloss over quite a bit.

“Delivery” implies that the aircraft will have to land somewhere or get low enough to drop an item without damage. That is likely to put it uncomfortably close to people who can be injured, or obstacles than can be damaged or cause damage to the unmanned aircraft itself. Those are sources of both safety and business risk that will have to be solved.

The economics of making such activity cost-effective pretty much demand multiple drones in the air at once under a single person’s control. A single delivery truck can drop off dozens of packages in a single run; the only way drones can match that is for them to be almost entirely self-guiding (autonomous), except perhaps at the very moment the delivery is taking place. However, even if a single pilot can supervise multiple aircraft and service each in turn for drop-offs, they all still would have to find their way to their delivery destination, and then home again. Right now, drones don’t have the technical means of doing that without being at risk of running into other aircraft or things sticking up from the ground.

Drones also can’t necessarily stay electronically connected to their pilots in congested, built-up areas over significant distances without expensive satellite connectivity. The proponents of delivery services don’t want to pay for that — it would make the proposition ruinously expensive over just using a truck — so they are exercising significant pressure on regulators (and legislators) to tell everyone else to stay out of their way so they can go make money with less liability and the ability to rely on autonomous aircraft that don’t need a human operator at all.

Say they get their way and the regulators get arm-twisted into allowing drone deliveries and similar activities at the expense of others. Some of the advocates for various commercial uses of drones are trying to walk a razor-thin wire, and I think it’s going to cut them sooner or later. In the U.S., the principle of “preemption” (The Supremacy Clause and the Doctrine of Preemption — FindLaw) allows them to say that they are operating under the FAA’s authority to regulate aviation operations. After all, people can’t sue anybody for flying an airliner 30,000 feet above their house. If a lower-level jurisdiction tries to ban drones flying within their city limits, as the linked article notes, “a federal court may require a state to stop certain behavior it believes interferes with, or is in conflict with, federal law.”

Now, say the operator wants to provide a service that is beneficial to those members of the community who want their sub sandwiches delivered fresh and hot, but annoys others who just want to get some sleep or don’t want to have to worry about their kids getting struck by a drone in flight. I can forsee such cases becoming common, every one of which could devolve into a vicious fight over the primacy of federal aviation regulations versus the rights of citizens in their own homes.

A canary of this type already has died. (Singer v. City of Newton-(Case Declaring Local Drone Law Illegal) I consider this an example of judicial overreach compounded by dumb legislation. However, the bald fact of the matter is that, in 2012, the Congress of the United States explicitly barred the FAA from making “any rule or regulation regarding a model aircraft, or an aircraft being developed as a model aircraft.” (P.L. 112–95, Section 336) There are a few special-interest qualifiers and weasel-words that follow this passage, but they’re essentially meaningless in actual practice. There’s too much room for misbehavior, and that latitude is going to mean that people who want to make money probably won’t have too much to worry about with respect to becoming public nuisances unless and until Congress reverses itself.

Based on this bad law, courts are now saying that preemption applies to aviation matters, so federal authority trumps that of the states; then they say that, since Congress said the regulator responsible for aviation matters — the FAA — may not make rules about model aircraft, they can pretty much do as they please and not be subject to registration, local ordinances or anything else needed for accountability or safety. Since you can’t tell drones in flight apart based on the intent of the operator — a business operator versus a “hobbyist” — all drones are going to start being viewed the same way by the public.

Getting away from grand strategy issues, let’s narrow the focus a bit. Your question alludes to urban congestion for the deliveries themselves. Your point is well taken. Using drones only to fly to central pick-up points — say, your local post office — would just be side-stepping the “last mile” problem. People want their pizzas delivered to their front door or their impulse buys plunked in their back yard, and the advocates for drone uses in such applications know that. Regardless of their “The Future Is Now” concepts, the long and the short of it is that, to me at least, drone deliveries to individual customers — especially in urban areas — are likely to be about as practical as flying cars, which similarly have been touted as the Next Big Thing in personal transportation literally for generations.

Delivery drones do make sense in niche applications where “last mile” concerns can be outweighed by virtue of the distance to be covered or the value of the cargo. There are highly worthwhile tests being flown using drones in other parts of the world to deliver medical supplies or other badly needed emergency items to isolated locations; it’s easy to see how beneficial such a capability would be in the wake of a natural disaster as well.

Still, clouds of drones bringing Stretch Armstrong collectibles from eBay to eager purchasers strikes me as a lot harder sell from both a regulatory and an economic perspective. The various sources of risk they might entail that have to be accounted for in the overall calculus of the delivery services themselves are non-trivial and complex. As a result, if customers aren’t willing to pay a serious premium for high-speed drone delivery, the business model seems likely to fail.

There are companies that have built or are building scores of central warehouse/distribution points all over the U.S. They use trucks or vans to take purchases to their customers… for the “last mile,” if you will. Since many of those facilities are in or adjacent to major metropolitan areas, surface transportation can do a pretty good job of satisfying consumer “I want it NOW” cravings within a few hours of an order being placed.

All of the above observations make me ask, “What’s the real value of even having a drone that can do the same thing as a truck if it’s going to cost more to do it, or anger the people affected by drone operations but are told they can’t do anything about them?” Honestly, the “last mile” factor suggests that drones may never be cost-effective. They make for great advertising on the basis of any given company’s commitment to the New and the Best and the Fastest. However, I personally believe that any brand that becomes interwoven with drone use is going to suffer brand harm every time somebody is hurt by one, or the noise from them becomes a source of community ire, or a drone is stolen by somebody with a big net.

Set aside the rhetoric, the promises of huge job growth and the pandering to public demands for instant gratification. Big-city use of delivery drones strikes me as a losing proposition. If would-be providers of such services can’t address all of the elements of the “last mile” problem satisfactorily with drones, and are perceived as creating community problems that outweigh their stated benefits, all it seems like they’d be doing is trying to create a high-tech identity that would become impossible to distinguish from any other user (or abuser) of drones every time something went wrong. Sometimes the simplest solutions are the best.

In the immortal words of Dennis Miller, “Of course, that’s just my opinion. I could be wrong.”

By Tom Farrier, Former Director of Safety, Air Transport Association. Originally published on Quora..

For more trending tech answers from Quora, visit HackerNoon.com/quora.