We've been following Zipline very closely for the last few years. The delivery drone startup has been operating in Rwanda since October of 2016, using small autonomous fixed-wing aircraft to paradrop critical blood products to rural medical clinics. The system is able to get blood from a centralized distribution center to where it's needed in minutes, independent of time of day, traffic, or weather. Zipline now manages 20 percent of rural Rwanda's blood supply, and has flown more than 300,000 kilometers (km) worth of commercial deliveries, carrying some 7,000 units of blood.
Today, Zipline is announcing major upgrades to its entire delivery system, introducing a bigger drone that can deliver blood faster and more efficiently than ever. The new hardware is already flying in Rwanda, and Zipline plans to bring their drones to the U.S. to demonstrate medical product delivery in suburban and rural areas later this year.
The new aircraft is part of a complete redesign of Zipline’s logistics system, which dramatically improves the system’s launch, autonomous flight, and landing capabilities. The improvements will decrease the amount of time between Zipline’s receipt of an order and launch of a fulfillment flight from 10 minutes to 1, increase the number of daily delivery flights that each Zipline distribution center can make from 50 to 500, and expand the radius of each distribution center to serve populations of up to 10 million people.
Zipline’s new delivery vehicle is an autonomous fixed-wing style airplane. The plane is capable of flying at a top speed of 128 km/h, and a cruising speed of 101 km/h—21 km/h faster than the previous generation of aircraft—with a round trip range of 160 kilometers carrying up to 1.75 kilos of cargo. The new plane is capable of flying four times faster than the average quadcopter drone and can serve an area 200 times as large.
While we won't go into an exhaustive analysis of what makes the new drone special (we'll save that for a detailed technical article, if we can swing it), CEO Keller Rinaudo did give us a few details about how much work Zipline put into the redesign. "Everything has changed," Rinaudo says. "We've spent the last year taking everything we learned from operating the system at national scale, and integrating that into a totally new platform. All of these iterations and improvements have come directly from serving patients, saving lives, and figuring out what those people need from a better system."
The most obvious difference is the speed of the drone (improved by 35 percent), as well as the improved range and larger payload. A lot of this is due to better aerodynamics— Rinaudo described the first generation drones as having the aerodynamics of a Humvee, and they managed to double the lift-to-drag ratio of the new drones.
Using fixed-wing drones is far more efficient than using rotorcraft, but it makes launches and landings particularly challenging. Zipline has been using a system of catapults for launching and an arresting system for landings, where the drone deploys a tailhook and snags a cable that brings it to a stop on a soft mat.
Those processes have been upgraded: the new launcher is much faster, and the deployable tailhook has been replaced with a tiny metal hook on the back of the drone's tail boom. This means that the landing target the drones have to hit went from about a meter in size to just one centimeter, but it reduces overall cost and complexity.
Making the drones easier and cheaper to build and to fix was another priority for Zipline. "It took hours to replace a part on the old aircraft," Rinaudo says, "so we put a huge amount of work into making sure this plane is very easy to maintain, and if you need to swap out a component, it's very easy and fast to do that." It takes about a tenth of the labor to build one of the new aircraft, and we're told they're "dramatically" less expensive, despite the larger size.
In addition to improvements to the drones themselves, Zipline has made what are arguably much more significant improvements to their overall system to try to reduce unnecessary delays. "The most obvious thing to us is speed is everything—when someone is waiting to receive an emergency medical product, minutes can be the difference between life and death," says Rinaudo. "We've optimized every part of the system, everything has been redesigned to allow us to go as fast as possible. One of the biggest problems was taking too long to launch vehicles, and taking too much labor to get a vehicle through all of the pre-flight checks and launched. That is the primary difference between doing 50 flights a day and 500."
Zipline is already expanding from Rwanda into Tanzania, and preparing to deliver a variety of medical supplies, not just blood products. The near-instant (or at least, very quick) delivery that Zipline offers has completely changed the supply chain in Rwanda, which in turn has a significant positive impact on the efficiency of the healthcare system overall.
"Rwanda has been able to dramatically decrease inventory at hospitals, which has reduced waste," explains Rinaudo. "Over the last year, they've been able to reduce waste at the hospitals that we serve to basically zero. They're the first country in the world to do that, and they've done it while increasing access to platelets and plasma by 162 percent. It's very promising how this technology could potentially [also] save tens or hundreds of millions of dollars in the U.S. healthcare system. Believe it or not, we have a lot of the same challenges here."
This may come as a bit of a surprise, since we tend to think of ourselves here in the United States as being comfortably advanced in all things, but this is the same sort of shift in thinking about how supply and demand work that has made Amazon so successful. The most practical way of distributing goods is to keep it all cheaply in a giant warehouse, and then give people exactly what they want very quickly. Amazon itself is (sort of) working on delivery drones for this reason, but Zipline has identified a much more immediately valuable niche with medical products. Medical facilities in the United States could potentially benefit just as much as Rwanda has, Rinaudo says:
We have a lot of the same rural healthcare challenges in the U.S. as you find in other countries. Rural hospitals are closing at a record rate, and people are having to travel farther and farther to get access to medical care. These are things that are fixable if we're willing to apply new technology, and Zipline feels very strongly that the technology is ready, and that it can operate now at national scale.
The key value proposition is that if you make instant delivery possible, you change the way that the supply chain works. You can keep more inventory centralized, and less inventory on hand. That means less waste, less carrying costs, and improvements in access to last mile primary healthcare clinics. If a patient comes in and needs a certain kind of treatment that the facility doesn't happen to have, there essentially is this portal, where they just reach their hand through the portal to the supply center and pull out the product that the patient needs and hands it to them.
We don't think that the promise of this technology over the next five years is to deliver pizza and tennis shoes to your backyard— we think the promise of this technology is to provide universal access to healthcare for every human on the planet.
By 7 May of this year, the FAA will announce participants in its UAS Integration Pilot Program, which is "an opportunity for state, local, and tribal governments to partner with private sector entities, such as UAS operators or manufacturers, to accelerate safe UAS integration." Essentially, the IPP will provide a structure under which companies like Zipline can (legally) operate unmanned commercial systems. While Zipline isn't making any specific announcements yet, Rinaudo did say that they're “working closely with a number of different states to demonstrate medical product delivery in suburban and rural areas."
To be clear, the idea is not to do one of those one-off delivery demos that we keep seeing. It's not that difficult to get temporary flight clearance to use a drone to make a point-to-point delivery over a relatively short distance under ideal (or at least good) conditions, and then put out a press release. Zipline is ready right now to operate at national scale, Rinaudo says, integrated with existing healthcare systems, making hundreds of long distance deliveries per day under just about any conditions. It may not be delivering pizza or tennis shoes, but it will almost certainly be saving lives, and establishing a precedent for how drone delivery can be a useful and reliable part of our healthcare infrastructure.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.