https://news.yahoo.com/look-sky-soup-155037137.html
Look, Up in the Sky! It’s a Can of Soup!

COLLEGE STATION, Texas — Exactly a decade ago, Amazon revealed a program that aimed to revolutionize shopping and shipping. Drones launched from a central hub would waft through the skies delivering just about everything anyone could need. They would be fast, innovative, ubiquitous — all the Amazon hallmarks.
...
Yet the venture as it currently exists is so underwhelming that Amazon can keep the drones in the air only by giving stuff away. Years of toil by top scientists and aviation specialists have yielded a program that flies Listerine Cool Mint Breath Strips or a can of Campbell’s Chunky Minestrone With Italian Sausage — but not both at once — to customers as gifts. If this is science fiction, it’s being played for laughs.

 

Will we see a transform from porch bandits to Drone strikers?

 

It's more pie-in-the-sky, "Oh, we'll all have flying cars some day," wishful thinking that makes no effort to consider reality.

This simply is not scalable, from a number of perspectives. Amazon delivers something like seven million packages a day. How large is their fleet of drones going to have to be to carry any worthwhile fraction of that?

Unless a drone can deliver multiple packages to multiple locations (and so far they can deliver a very small package to one location -- but that would change), they are competing with normal delivery methods that are highly bundled -- a driver would be expected to deliver something like 100 packages in a shift and perform pick-ups at the same time. So the drone has to get loaded for one delivery, fly to the deliver site, drop it off, and fly back for another load for less cost than a driver going from one stop to the next. Time is another factor. You have to have enough drones to make those 100 deliveries in about the same time as that driver, otherwise you will just get progressively behind. And then you still have deal with pickups. If that means sending out drivers to do the pickups, then a good fraction of the deliveries they make are being piggybacked onto something that still has to be done, and so the marginal cost is greatly reduced.

But perhaps the biggest thing that all of these mass-drone-use folks conveniently ignore -- the same as the flying-car proponents -- is that you simply cannot have that many aircraft in a congested airspace without extremely tightly planned and coordinated traffic control. Over the entire United States, the daily aircraft traffic (commercial, civilian, business, government, and military) peaks around 16,000. Much of that traffic is on coordinated instrument routes and schedules that were worked out months and years in advance.

Is there a potential niche for drone deliveries? I think so. There are probably enough rural deliveries where you currently have to have a vehicle and driver dedicated to a single delivery for a significant amount of time to justify having a small fleet to service those deliveries. The same thing happens now with air travel -- in remote locations, there are small companies that specialize in passenger and cargo operations on a scale that is sustainable by that region, but it's the small companies, by and large, that can make a go of it. Small companies are flexible and adaptable whereas the big companies tend to insist that all operations fit in a single, large-scale model that just doesn't apply to those situations.