The biggest limiting factor in the article appears to be the five-pound limit. If that's going to persist, I don't know why they're even piloting it, other than branding (also mentioned in the article) and stubbornness. Inherent efficiency deficit vs. wheels or tracks.
(Second biggest limiting factor - weather.)
Third limiting factor - everyone has a shotgun.
I think that's the funniest NYT article I've ever read.
If you pay for the peanut butter, can you have it dropped into your neighbor's skylight?
Are you trying to give them a pill?
Moving something heavy through the air is just always going to be ridiculously inefficient, which is going to limit its applications to light-weight, high value transactions. Sending cans of soup through the air is silly. Delivering weed, on the other hand....
I want someone to run the numbers on the relative costs of this drone project versus installing pneumatic tubes to people's houses. Subs and burritos for everyone!
9: Retvrn to tradition!
https://en.wikipedia.org/wiki/Pneumatic_tube_mail_in_New_York_City
UPMC just built a tube system for medical specimens and the like.
8 So how long after weed starts getting delivered do we end up having predator drones that can capture a weed drone and drag it off course? This would seem like a fairly easy step for someone to take.
14: Since they deliver from only 12 feet up, who needs a drone? Bring back the retiarii.
Its the one-way, cardboard drones carrying 6.6 pounds of C4 you got to watch out for.
I can imagine a lot of places where those drones will be just free skeet shooting targets.
When I worked at NYPL they still had the pneumatic tube system for sending call slips in place though it was no longer in use.
"Moving something heavy through the air is just always going to be ridiculously inefficient, which is going to limit its applications to light-weight, high value transactions. "
I don't know about that- the alternative, remember, is moving it along the ground in a huge two-ton steel vehicle burning fossil fuel. Drones are very small and light.
Now, Amazon gets past this by putting lots of parcels in the same van. Motorcycle couriers don't, though, and it's very possible that this might be a lot more efficient, and certainly less polluting, than a motorbike.
Let's run some numbers!
Amazon expects its drivers to deliver 40 packages in a three-hour block. That makes sense - forty average-sized Amazon boxes take up 3.4 cubic metres, which is about the cargo volume of a medium sized delivery van. So a three-hour block equates to delivering a vanload of packages.
During that time you'll drive about 75 miles, constantly stopping and starting - which makes for terrible fuel efficiency. Say 10 mpg.
So, 40 packages for 7.5 gallons of fuel - or 280 kWh in energy terms. 7 kWh per delivery.
A motorcycle courier, flat out, will do about four urban deliveries an hour, in which time they'll travel 10 miles (normal urban traffic speeds) at 40 mpg. So, motorcycles will deliver those 40 packages in 10 hours, travelling 100 miles, for 2.5 gallons of fuel - 113 kWh. 2.8kWh per delivery.
That's interesting. Motorcycle couriers are much more fuel efficient than delivery vans! Even though they travel further for the same number of deliveries. That makes sense because they can't carry 40 packages at once - they're going back to HQ every time, or, at best, every two or three times.
What they aren't, of course, is more time-efficient - you're burning less fuel but more man-hours.
Now, what about drones?
Amazon doesn't publish drone specs and the NY Times, being a joke newspaper, doesn't bother to look into them because numbers are hard. But a cargo drone with a payload of 2kg typically operates for about half an hour on a battery pack of 8,000mAh at a voltage of 22V - in other words, 0.2kWh. (Energy equals voltage times current times duration).
How many deliveries could it make in that time, at a cruising speed of 40-50mph? Well, drones are definitely going back to HQ after each dropoff - a motorcycle courier might be, or he might manage two or three at a time. Not more, because we're talking about standard Amazon boxes. Unlike motorcycle couriers, though, drones can travel in straight lines. Let's say it nets out and a drone can manage 40 deliveries in 100 miles, just like a courier. It's travelling four times as fast as the biker, though, so it'll be done in 2.5 hours. Call it three hours. (I'm ignoring recharging time because we are talking about energy efficiency here - so deliveries per kWh, not deliveries per drone-hour.)
So the final tally looks like this:
Delivery van - delivers 40 packages in 3 hours for 280kWh
Motorbike - delivers 40 packages in 10 hours for 113 kWh
Drone - delivers 40 packages in 2.5 hours for 1.2 kWh.
Unlike motorcycle couriers, though, drones can travel in straight lines.
Can they, though? A world in which drone delivery is more than a novelty is surely one in which delivery drones are routed via pseduo-ATC to avoid a totally chaotic airspace.
21: fair point. But that doesn't alter the final result - even if we assume that drones travel triple the distance of motorbike couriers per delivery, we're still talking about three kWh vs 113 kWh for that forty-delivery shift. In energy efficiency terms the drone advantage is so massive that it only gets overturned if I've made some truly heroic mistakes (like confusing watts for kilowatts or something). Which is possible?
The real disadvantage to drones is probably capital cost. To keep one drone in the air you probably need another four on the ground recharging or being maintained. That's a lot of capital, compared to one dude on a motorbike or even one dude in a Transit van. Those drones are about $3-4k each, and I imagine they aren't nearly as durable as a Transit van.
And the other disadvantage, of course, is that a man in a van can carry bigger parcels if he wants. I've run the numbers using standard Amazon book boxes, but of course I can order 15kg of dog food from Amazon if I like, or a lawn mower, or a flatscreen TV, or a king-size mattress. The drone couldn't handle those. So drones are, apparently, vastly more efficient, but less flexible.
I think where people are going wrong on this point is that they are breezily asserting "well, air transport is much less efficient than surface transport" as if this was an inherent fact about all surface transportation.
It's true when you're comparing cargo aircraft to freight rail or sea freight, of course. But they aren't stopping to think about whether a Transit van making its way through city streets is any different from a mile-long freight train on steel rails, or a 20,000 TEU freighter crossing the Pacific.
Swallows would be even more energy efficient than drones.
Prof Stormcrow! Why is yhe E coast of Madagascar so straight?
Not completely sure why it is *so* straight. But will come back to that after going through what seems to be the origin story. Spoiler: The relatively straight SW coast of India is basically its counterpart.
Madagascar split from Africa as part of the breakup of Gondawanaland during which it was still attached to India. Some tens of millions of years later India split off (along those straight coasts) and powered up into Asia. The first split was early Cretaceous , the latter late-ish Cretaceaous (~150Mya and 80Mya). The volcanism that created the ones you first noticed are from this time period. It is slightly more complicated as there is the ridge (not a spreading ridge but a remnant) of which the Seychelles are the above sea level expression (but I think that was attached more at the northern part where it is less straight).
The Madagascar/India split seems to be mostly a strike/slip or transform (think San Andreas) rift than a mid-ocean ridge like in the Atlantic or now in east Africa. They tend to be straighter (look at the little east/west transform faults in the mid-Atlantic that connect the spreading ridges) and that makes sense mechanically when you think of relatively solid things sliding past each other. They can have broad bends (San Andreas in SoCal for instance) but this segment is a relatively straight one (and may have had some spreading component as well as the lateral motion).
As to why the different splits happened when and where, seems to be related to the comings and goings of hot spots (plumes in the mantle).
Not clear if Madagascar is still a separate "plate;" it seems to have been moving in conjunction with the Somali plate (east side of African rift) for a while now (bit old plate boundaries can be dormant for a long time and then "re-activate"). There is a chance Somalia and Madagascar will plow into India now that it is stuck up against the Himalayas. But the what is going on in the central Indian Ocean seems complicated. Prediction sabout the future you will not live to see are simultaneously hard and very easy.
I've never seen a motorbike delivery service. Probably because of density.
I think you're flying those drones pretty fast.
If the drone can't deliver when the weather is bad, you've got a lot of drones sitting in your shop doing nothing, while your drivers (whom you're still employing) are driving vans (which you still own and maintain).
Drone delivery to single family suburban homes are pretty different from drones to dense cities with lots of apartment buildings. If 7 different drones are making 7 different deliveries around the same time to seven different customers at the same apartment building, you'll want someone on the ground making sure everything gets to the right person.
Eventually, you'll be replacing your fossil-fuel vans with electrics.
Aren't couriers in big cities on e-bikes nowadays anyway?
Big enough buildings have doormen or mailrooms. All buildings have windowsills on which I'm sure many well-funded startups are eager to bolt IoTed blockchained etc mailboxes.
Charlie Munger wants to take away the proletariat's windows.
33 Yes, but instead of a once a day delivery of the building's deliveries to the building, you're dribbling it in all day long. And because the point is 30 minute soup delivery, the building staff is calling the resident to the mailroom. This is the building's problem not Amazon's, but basically Amazon is dragooning building staff.
I've never seen a motorbike delivery service. Probably because of density.
Well, not entirely surprising given where you live. They exist in big cities, though.
I think you're flying those drones pretty fast.
Based on the manufacturer's webpage. But, sure, cut the speed to 10mph if you want. That means they'll complete those 40 deliveries in ten hours, which is 20 recharges of 0.2kWh or 4 kWh. Still vastly more efficient than a van or a bike.
Eventually, you'll be replacing your fossil-fuel vans with electrics.
Which will still be far less efficient than drones, because they are still large, heavy things that need to be moved around - though they'll be cleaner than ICE vans.
The big issues are going to be weather (as you say) and also possibly control range. Longer range means more powerful radio links. Though Amazon seems to have solved that for Texan sprawl and, as you say, this is a sprawl solution, not a dense urban core solution.
Aren't couriers in big cities on e-bikes nowadays anyway?
Some are. And an e-bike is close-ish to drone performance. You'd need three charges at least to cover that 100-mile delivery route - each charge is 24Ah at 60V = 1.5kWh.
Because the NY Times is a joke newspaper, it prints things like this:
The Texas weather plays havoc with important deliveries. Mr. Lord, a 54-year-old professor of civil engineering at Texas A&M, ordered a medication through the mail. By the time he retrieved the package, the drug had melted. He's hopeful that the drones can eventually handle problems like this.
Really not seeing how this is a drone problem. If you have a vital temperature-sensitive medicine to deliver, then leaving it sitting around in the summer in Texas is probably not a bright idea, however it is delivered to your door.
And the other disadvantage, of course, is that a man in a van can carry bigger parcels if he wants. I've run the numbers using standard Amazon book boxes, but of course I can order 15kg of dog food from Amazon if I like, or a lawn mower, or a flatscreen TV, or a king-size mattress. The drone couldn't handle those. So drones are, apparently, vastly more efficient, but less flexible.
So how about calculating per kWh per kg?
The U.K. is in the middle of a diarrhea outbreak. People should be washing their hands instead of doing math.
This is true. Crom currently has both diarrhoea and an ear infection. She is runny at both ends.
Interesting analysis. I would add the complication that when Amazon ships me three things, they often use one box, and now they may need to use 3 boxes, making for three trips instead of one.
When I worked at NYPL they still had the pneumatic tube system for sending call slips in place though it was no longer in use.
I think the Library of Congress was still using some of the pneumatic tube system in the main reading room as recently as the early 2000s.
Someone run the numbers on converting Unfogged to a pneumatic tube system, just to see.
The whole internet is pneumatic. That's what the fan in your computer is for.