When a fire broke out in a slum on the edge of Cape Town, South Africa, on New Years Day in 2013, it took firefighters six hours to stop the blaze as it spread from shack to shack. By the time the fire stopped, 4,000 people were homeless and four people had died.
Apart from the scale, the tragedy wasn’t unusual. Just four days earlier, another eight people died in a nearby fire. By some estimates, in South African settlements, there are as many as 10 fires every day.
A group of more than 20 partners, including the Red Cross and the design firm Frog, is piloting a new community-wide system designed to help prevent fires or catch them before they spread across a neighborhood.
The challenge isn’t simple to solve. Cheap shacks are packed closely together, so fires spread quickly. Shacks don’t have street addresses, so they’re hard for fire trucks to locate, and even if a firefighters know where to go, streets are so narrow and crowded that it’s often impossible to drive there. Because people cook on open fires, use lanterns and candles, and often hack together illegal electrical wires, the risk of fire as high.
The team started by testing a low-cost sensor from a startup called Lumkani, developed by engineers at the University of South Africa. Unlike a typical fire sensor, it doesn’t look for smoke. “Because there is cooking on indoor fires, if you were detecting smoke they would be going off multiple times a day, and it would be quite useless,” says Julie Arrighi, a resilience advisor at the American Red Cross. “So these are actually detecting rate and rise of heat.”
The sensors are also networked, so neighbors immediately learn if there’s a fire in a nearby shack. “The proximity of the homes make it so that fire is actually a community problem, not an individual problem,” says Fabio Sergio, global lead of Frog’s social impact practice. “It’s actually just as dangerous to have your neighbor’s house burning down, because very quickly it becomes your house burning down too.”
If a sensor detects a fire, it can send a text message to neighbors. If someone has children home alone–something that’s common if a family can’t afford daycare–they can call a neighbor to rescue them. Those who are nearby can respond with buckets of water. The sensor can also send a GPS location to official fire trucks.
“In the end, our design is setting up more than a service and less of a detection system,” says Sergio. Along with the sensors, the team is testing the idea of a “Fire Club” of residents trained in fire prevention and running community response stations. They’re mapping out locations of water sources, and figuring out how to bring firefighting equipment closer to homes so they can be reached in time.
They’re also figuring out how to make the whole service into something that’s cheap enough that the community can afford it. It’s intended to be run by local micro-entrepreneurs, not provided by outside humanitarian organizations.
“If the device is at the right price point and people aren’t willing to buy it, then we haven’t really met their needs,” says Arrighi. “Ultimately, too, there’s a scale issue involved. Globally there are over 800 million people living in informal settlements around the world…those though the Red Cross is the largest humanitarian organization in the world, it’s still a very big scale, and we can’t do it alone.”