An Algorithm Is Helping a Community Detect Lead Pipes

More than six Years after residents of Flint, Michigan suffered widespread lead poisoning from their drinking water, hundreds of millions of dollars were spent improving water quality and fueling the city’s economy. But residents still report some type of PTSD in the community, waiting in long rows of grocery stores to get bottled water and filters. Media reports on Wednesday said former Governor Rick Snyder had been charged with negligence for his role in the crisis.

Snyder maintains his innocence but told Congress in 2016: “Local, state and federal officials – we have all abandoned Flint’s families.”

One tool that emerged from the crisis is some form of artificial intelligence that could prevent similar problems in other cities where lead poisoning is a serious problem. BlueConduit, an analytics startup that claims to use predictive models to find lead pipes, had promising results in Flint, but the city’s complex politics ended their use prematurely.

Now, four years later and 100 miles away, officials in Toledo, Ohio who have concerns about lead pipes are planning to use the technology. They hope to avoid the issues that popped up in Flint by expanding the reach and engagement of the community. The Ohio Department of Health estimates that 19,000 children in the state have elevated lead levels. Children in Toledo tested almost twice as positive for lead poisoning as nationwide, according to a 2016 report by the Toledo Lead Poisoning Prevention Coalition.

Lead is a debilitating neurotoxin that can cause lifelong development problems in children and is toxic to adults even with low exposure. Last year, Toledo committed to a 30-year project to find and replace the estimated 30,000 lead pipes in the city. In October, a coalition of the city, local activists, and a nonprofit group received a $ 200,000 grant from the Environmental Protection Agency to use BlueConduit technology to find lead pipes.

Founded in 2019 by Jacob Abernethy and Eric Schwartz, BlueConduit emerged from a University of Michigan project to identify lead pipes in Flint. According to Abernethy, the startup has contracts with organizations that govern 50 cities to support lead pipe replacement.

“We say: Here is the ranking of the probabilities. And if your goal is to reduce the time people in the community live with lead pipes, this is how you should go through the list. “

Eric Schwartz, BlueConduit

BlueConduit uses statistical techniques to predict which neighborhoods and households are most likely to have lead pipes based on dozen of factors: the age of the home, the neighborhood, the proximity of other homes where lead has been found, utility records, and more. With a list of addresses, BlueConduit offers a ranking based on the likelihood of a lead service line. Cities can use the ranking to prioritize houses for excavation in order to examine the pipes.

“You can imagine that less than” These houses have lead, these houses don’t, “” says Schwartz. “We say here is the ranking of the probabilities. And if your goal is to reduce the time people in the community live with lead pipes, this is how you should go through the list. “

Alexis Smith, community program and technical staff member at Freshwater Future, a nonprofit that works with Toledo, says one of the attractions of Toledo is the input from residents and the algorithms.

“It gets the knowledge of the homeowner and information not only from the city but also from the residents,” she says. “We were really reassured that this wasn’t just going to happen to us. We will work as part of this program. “

The balance between technology and community perspectives is important so that residents don’t feel that their concerns are subordinate to the algorithms. During the Flint project, BlueConduit’s model produced promising results, but encountered a divided community and deep distrust of leadership.

In 2017, Schwartz and Abernethy, professors of marketing and computer science, worked with Flint representatives who were initially impressed with the team’s predictive model. This year around 70 percent of the homes identified by the model had lead pipes. The city later signed a contract with AECOM, a Los Angeles-based engineering firm, which refused to use the couple’s predictive models. Without the model, the accuracy dropped to around 15 percent the following year.

You might also like

Comments are closed.