Zipline’s Ambitious Medical Drone Delivery in Africa
You can hear the drone before it’s visible, whining like a mosquito above the hillside grounds of Rwanda’s Kabgayi District Hospital. Emerging through a patch of fog, roughly 100 feet in the air, the small plane quickly disappears again, circling in an oblong pattern as it descends toward an altitude low enough to make its drop. After a period of silence, it’s suddenly back, swooping over the roof of Kabgayi’s accident ward to drop its payload on the driveway with a thud. On the ground lies a red cardboard container, roughly the size of a shoebox, attached to a parachute made of wax paper and biodegradable tape. The contraption may resemble a children’s art project, but its contents are lifesaving. Packed tightly inside are two units of human blood, which will probably soon be used for transfusions during surgeries or complicated childbirths, or to treat young victims of malaria.
The plastic sachets of blood are among the first commercial products ever delivered by drone, part of a partnership between the Rwandan government and the Silicon Valley–based robotics firm Zipline, which began introducing the blood drops at Kabgayi in late 2016. The service, which is now delivering to seven of 21 planned facilities, is still in its infancy. Yet it has already had an impact. In the past, hospital staff would make three drives per week to procure blood products in the capital, Kigali, 60 kilometers away, a three- to four-hour round trip. Emergencies meant additional trips—sometimes resulting in life-threatening delays. Now, a Kabgayi lab technician simply taps out an order on a smartphone and Zipline’s distribution center, located five kilometers from the hospital, will have a drone there within 15 minutes. “Before, it was a serious problem to have blood when we needed it,” says Espoir Kajibwami, a surgeon who served as Kabgayi’s medical director until February. In emergency cases, he says, the hospital would often send the patient to the national referral hospital in Kigali rather than wait for the blood to arrive.
Zipline’s blood deliveries come at a time of great activity in the world of drone-enabled commerce. Last December, three years after announcing its much-publicized Prime Air service, Amazon conducted its first commercial drone delivery, to a farmhouse in rural England. The month before, the convenience-store chain 7-Eleven completed 77 on-demand drone deliveries of pizza, Slurpees, and over-the-counter medicines to customers in Reno, Nevada. UPS, which has helped finance Zipline’s operations through a $1.1 million grant from its charitable foundation, delivered a package in February with a drone launched from the top of one of its signature brown trucks. Flirtey, the drone maker behind the 7-Eleven pilot, has also tested the delivery of medicines in understocked parts of Appalachia. Another U.S. firm, Matternet, has conducted test flights in collaboration with UNICEF to deliver infant HIV test kits in Malawi.
Yet Zipline, which uses fixed-wing drones that have a greater range and are more resilient in bad weather than the more common multicopter models, is the first in the world to offer regular delivery of emergency medical products.
Founded in 2011 under the name Romotive, the company first gained notoriety as the maker of Romo, an iPhone-powered robotic pet, before CEO Keller Rinaudo decided to seek a product that would have a greater social impact. Soon he and Zipline cofounders William Hetzler and Keenan Wyrobek were scouring the developing world to learn how drone-based logistics could help save lives.
On two separate trips to Tanzania in 2014, Rinaudo and Hetzler, who’d first worked together as Harvard freshmen to build a rock-climbing gym on campus, met a researcher working with a local NGO who’d created a text-message-based health surveillance system. Through a network of community health workers, the project tracked hundreds of health emergencies across the country, including snakebites, possible rabies cases, and severe postpartum hemorrhaging. Frequently, however, reaching remote patients with treatment in time was cost-prohibitive or logistically impossible. “You have this database full of tragic human stories,” says Hetzler, 30. “The obvious missing piece was a way to very rapidly respond to that demand and get the product into a place where it isn’t otherwise readily accessible.” The pair came away from their visits feeling they could build that.
Under the technical direction of Wyrobek, Zipline’s engineers got to work at the company’s headquarters in Half Moon Bay, California, developing a craft with twin electric motors that could carry a 1.5-kilogram payload and operate in almost any weather. As the technology progressed, Rwanda emerged as an ideal spot to test the Zipline vision. Similar to Tanzania, its neighbor to the east, the small East African country is home to a predominantly rural population that is plagued with transportation challenges. A majority of Rwanda’s 478 health centers, and many of its 35 district hospitals, can be accessed only via poorly maintained unpaved roads, which often spiral into its famed “thousand hills” and are difficult to traverse by vehicle, particularly during the twice-yearly rainy season. Unlike Tanzania, though, Rwanda is compact: with 12 million people in an area the size of Maryland, it is the most densely populated country in mainland Africa. This meant that Zipline’s drones, which have a flight range of 150 kilometers, could serve nearly half the country from a single launch site.
Zipline’s idea caught on with Rwandan authorities, including the country’s civil aviation body, which altered regulations to enable its drones to operate. In mid-2016, Zipline signed a deal with the Rwandan government to build a distribution center near the town of Muhanga.
This hilltop site, erected on a former maize field, is now known as the nest: a fenced-in plot consisting of a white circus-style tent housing a blood storage facility, 13 drones (nicknamed “zips”), and a small staff of young Americans and Rwandans. On one side of the tent, two stainless-steel launchers, facing opposite directions to account for changing winds, employ a system of bungee pulleys to catapult the 12-kilogram drones into the air at 84 kilometers per hour. On the other, two brown inflatable mats cushion the zips’ landing on return. Once the drones are airborne, cruising over an undulating landscape dotted with banana trees, cassava fields, and tin-roofed houses, an operator monitors their path on an iPad, staying in constant touch with air traffic control in Kigali. All routes, developed using a 3-D satellite map followed by detailed manual ground surveys, are pre-programmed using real-time kinematic satellite navigation, which—along with an inertial navigation system—enables the payload to drop within a target area five meters in diameter. “Accuracy is extremely important,” says Hetzler, adding that if the drops were less precise, packages could end up on roofs, in trees, or in other inaccessible spots that could destroy the operation. He says the company is developing technology that will automate the ground surveying process.
Zipline’s plans for Rwanda include scaling up to a wide range of medical products, including emergency rabies vaccines; drugs to treat HIV, tuberculosis, and malaria; contraceptives; and diagnostic test kits.
Yet blood represented a natural starting point. After all, it has a shelf life of only 42 days, must be kept refrigerated, and is frequently needed on an urgent basis.
Today, Rwanda’s Ministry of Health stores blood at a national center in Kigali and four regional depots around the country; its 58 facilities equipped to handle blood transfusions, mainly hospitals, keep a small inventory of common blood types and must continually restock from the depots or national center. Sometimes the stock on hand is sufficient for emergency transfusions. Often it is not, particularly if the patient has a less common blood type. In these cases, the facility must refer the patient or collect the blood by car, motorbike, or truck.
Among the individuals most vulnerable to blood delivery delays are pregnant women. Although Rwanda’s maternal mortality rate has declined by more than two-thirds since 2000, an improvement linked to increases in contraceptive use and a move away from home births, complications of childbirth remain a leading cause of death. The World Health Organization estimates that Rwanda has one maternal death for every 344 live births, 20 times the rate in the United States and 97 times the rate in the top-performing countries in Europe. More than half of maternal deaths occur after childbirth, and 26 percent are the result of hemorrhaging. Faster, more reliable access to blood could help reduce this number, along with lives lost to accidents and malaria-induced anemia, which is common in small children.
Kabgayi, one of Rwanda’s largest district hospitals, faces a higher blood demand than most. The colonial-era facility, several aging brick buildings on the site of a former Catholic mission, uses up to 100 units per month, according to Kajibwami, the former medical director. Roughly half of that demand comes from the maternity ward, which recorded more than 4,600 births in 2016. The hospital’s high volume of patients was evident on a February morning from the crowd that sat on wooden benches in its open-air reception area: women in colorful patterned dresses, men in fraying jackets, uniformed schoolchildren, and even a small group of prisoners dressed in pink and orange smocks.
Although many Kabgayi patients come from hard-to-reach surrounding villages, the hospital itself, which sits just off a well-maintained paved road to Kigali, isn’t difficult to access. The facility is also too close to Zipline’s distribution center for drone deliveries to make much sense absent plans to scale up to other hospitals: if its drones failed to function properly, the company reckoned when selecting Kabgayi as its inaugural site, it could still deliver blood quickly by truck. Yet most of the facilities Zipline plans to service are considerably more remote. Personnel at Muhororo District Hospital—Zipline’s second pilot facility, which received its first drone delivery at the end of February—could spend an entire day battling rutted dirt roads, and the occasional landslide, to make a blood collection in Kigali.
Zipline’s impact is hard to quantify. Difficulty accessing blood is not the only factor inhibiting effective care: physicians and other skilled personnel are in short supply, and many patients have trouble reaching medical facilities. A majority of pregnant women plan to deliver at less equipped, more remote health facilities that typically lack doctors or blood transfusion capabilities. Complicated births, therefore, are usually referred to hospitals—a process that can cause critical delays in treatment for which drones have no answer.
Few inside Rwanda, a highly authoritarian state, dare to openly criticize the government or its programs. Yet many who’ve heard about the project are skeptical in private and question why authorities are investing in unproven high-tech gizmos when demand for fundamentals like ambulances and health workers still exceeds supply. To some critics, the project’s most important benefit could be the good PR it generates, which distracts from the Rwandan government’s spotty human rights record and the acute levels of poverty in much of the country. Rwanda, which is highly reliant on foreign aid, has long marketed itself as a regional hub of tech and innovation—an idea that remains more of an aspiration, and a strategy for generating outside interest, than a reflection of current reality.
It doesn’t help that the project’s costs are not disclosed. Rwanda’s government pays Zipline a fixed price per delivery, with a minimum volume guarantee, but neither will divulge the terms. Hetzler admits that routine drone deliveries for restocking are, for now, more expensive than standard trips by road, but he says that his drones have cut costs in emergencies. In addition, he argues, routine costs will be offset if Zipline makes possible a far more agile supply chain—for blood as well as more durable products like medicines—in which fewer items must be kept at last-mile facilities, minimizing waste and more effectively preventing stock-outs. And as a general rule, per-delivery costs should decrease as flight volume expands. Bruce Y. Lee, a professor at the Johns Hopkins Bloomberg School of Public Health who led a research team last year that simulated a drone-based vaccine supply chain in Mozambique, says drones “certainly have the potential to both decrease costs and increase demand fulfillment,” depending on the frequency of flights, the distances traveled, and the speed of trips on local roads.
In Zipline’s case, better logistics also mean that facilities can access products with shorter shelf lives or unique storage requirements. In addition to blood units of all types, Zipline’s existing service offers platelets, fresh frozen plasma, and cryoprecipitate, all of which promote blood clotting and were previously underutilized because they were too difficult to store at health facilities.
A fuller picture of Zipline’s impact on Rwanda will probably not take shape for some time. After commencing operations at Muhororo and five additional sites, the company is working to begin deliveries at the 14 other planned facilities by the end of this year. In addition to introducing new products, Zipline plans to build a second distribution center in the country’s east. This, together with future-generation drones expected to have longer ranges, would enable it to cover all of Rwanda except Kigali, where products are stocked in sufficient quantities already.
Additional countries are also in the pipeline. According to Hetzler, a “long list” of governments have expressed interest in Zipline’s technology, and the company has already etched agreements with Tanzania’s Ministry of Defense and Civil Aviation Authority that would enable it to launch there next. To finance its expansion, Zipline raised $25 million last summer, bringing its total capital raised to $43 million. Still, it is choosing its next locations carefully, with a priority given to countries that offer a chance for social impact and are willing to make necessary regulatory changes and a firm financial commitment. Eventually, Zipline also aims to operate in wealthier countries, including the United States, where regulations currently restrict drone flight to within an operator’s visual line of sight absent special review.
Back at Kabgayi Hospital, health workers say the service has already helped. Inside the modern laboratory building, where blood is refrigerated in a small, ceramic-tiled room, Prosper Uzabakiriho, a white-coated lab technician, says that cutting back on lengthy road trips to Kigali has allowed many patients to be treated more promptly and freed up time for staff to perform their duties. At the time of my visit in mid-February, the facility had received an average of 11 deliveries per week since the start of the year and was entirely reliant on Zipline for blood products during normal business hours. By late May that had climbed to 20 per week, including weekend deliveries. Rwanda’s civil aviation authority has not yet given it the go-ahead to operate at night, so Kabgayi will continue to rely on its former blood supplier in Kigali for emergencies after hours. (According to the Ministry of Health, the country’s five existing blood depots, which also serve as centers for blood donation, will remain in operation even as Zipline scales up.)
The surgeon Kajibwami credits the service with helping him respond to one particularly urgent case: a woman who began hemorrhaging after surgery to remove an ectopic pregnancy. The ensuing operation, which entailed a partial hysterectomy, required six units of blood, more than Kabgayi had in stock, and thus an urgent call was made to Zipline. The doctor cannot say whether the patient would have died in other circumstances. Still, he’s grateful for the timely delivery. “It was very bad,” he says of her condition. “It would have been very difficult to manage it.”
How to preserve your digital memories
Following recent announcements by Google and Twitter, more data deletion policies are coming.
Your digital life isn’t as permanent as you think it is
Google will delete accounts after two years of inactivity, and experts expect more data deletion policies to come
Catching bad content in the age of AI
Why haven’t tech companies improved at content moderation?
Behind the scenes of Carnegie Mellon’s heated privacy dispute
Researchers at Carnegie Mellon University wanted to create a privacy-preserving smart sensor. They were accused of violating privacy instead.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.