AI/Watson

WaterBot: How to know if your water is safe to drink

Share this post:

How can you tell if your water is safe to drink?

Most Americans worry about their drinking water, with good reason. News reports regularly expose high levels of PFCs, PFASs and other contaminants found around the nation. In 2014, we heard that the drinking water in Flint, Michigan, was discolored with dangerous contamination. Over the next two years, investigations unearthed a scandal and charged nine officials with hiding evidence of contaminants and the resulting spike in lead poisoning among children.

In the aftermath, I served on a project for the Water Quality Association of Wisconsin addressing similar lead concerns. But I realized that our project was only going to check the water once or twice a week. I wanted to go further. While weekly tests can identify slow trends, they might not alert people to the immediate effects of an industrial spill, local system breach or other failure. I thought there had to be a faster way to check water quality. That’s where WaterBot began.

A data revelation

I worked with a partner, Edgar Duarte, to build a small device that monitored the water quality in a home. That’s important, and we developed incredibly accurate monitors.

But then, Edgar had an idea. Looking at the water quality data, he said, “We’re sitting on incredibly powerful information… if we crowdsource it.” He envisioned users anonymously sending their data into a central system that would effectively track water quality across the nation. The system could help catch a crisis by providing real-time analysis and absolute transparency apart from any utility or government body.

That was the moment when we were both inspired by the project—and terrified about the massive quantity of data to be collected.

Remembering the allegations in Flint, we also realized that some people might want to stop this information. Our data needed to be transparent. We needed to crowdsource it to an open source database in the public domain.

That was a revelation for the project.

Connecting the water

Initially, we connected our devices into a Google cloud with a NoSQL framework. We started to see results, like an alert that came in when a waterline’s debris level jumped by 40 percent within minutes—we later found that there was a breach at a nearby construction site. But we needed more scale, more data and more analysis to diagnose problems remotely.

I was at a conference in Reno when I received an alert about the municipal water system in Waukesha, Wisconsin. I knew there was a problem, but I didn’t have enough information to diagnose it. About 20 minutes after the alert, I received a call about the IBM Watson IoT platform. It was the perfect moment.

The Watson IoT platform gave us the power of scale, and our vision for a transparent water quality database was on the horizon. Watson allows us to quickly build a robust, real-time algorithm and update it as needed. To scale up our device production, we connected to the partnership between Indiegogo crowdfunding and Arrow manufacturing.

The people at Indiegogo were a valuable source of support for us. And the community there helped give us validation and confidence about our vision. Just to have that social proof—that validation—made those nights of trying to get everything off the ground seem a little less long.

Clear confidence

Now, we’re about to see this vision become a reality in the next 30 days as these first devices ship.  The Open Water Quality Project, owq.org, will stream that data.  To keep the data completely secure, we will be using blockchain technology. People often associate blockchain with financial transactions—for us, it helps to ensure that water quality results cannot be intercepted and changed by anyone.

I think blockchain is absolutely critical to the process of restoring trust, particularly in the public sector.

A joint venture between the public sector and the private sector, combining resources with secure transparency, could be a huge step towards restoring some of the trust that’s been eroded.

Ultimately, we want to unify civic and private water quality monitors. We hope to use IBM Smarter Cities technology to assemble citywide grids of civic and private sensors into early warning systems that could forecast changes, generate alerts, locate problems and diagnose causes. It could keep citizens safer and save cities from tearing up roads to find an issue. I believe we can radically alter our approach and help ensure that we can intercept the next incident like the one in Flint.

The groundswell of support that WaterBot has seen has already been so rewarding. It’s exciting to see the genuine feedback from people who are willing to crowdsource data to help provide a safer water-management grid for everyone.

We need to continue to shine more light on this problem. We need greater transparency. We must make sure that the future of our water is clear.

  

For more, read the IBM Watson IoT blog or watch the IBM interview with Chris Richter below.

 

Co-Founder

More AI/Watson stories

How AI chatbot technology can help boost service

What makes a financial services organization a good bet for a customer? To start, customers look for a successful track record, a commitment to fiduciary duties and low fees. But they also value responsiveness. When customers have a question, they want a prompt answer. They want a company devoted to providing convenient, personalized customer experience. […]

Continue reading

How to find new football stars with AI technology

For football (or soccer, for American) fans, talent on the field is a thing of beauty. Names like Messi, Ronaldo, Pelé and Zidane evoke images of unparalleled grace and athleticism. As we saw in this year’s World Cup, new stars are emerging all the time. But have you ever wondered how new talent rises to […]

Continue reading

How AI technology is helping Lufthansa soar to new heights

Digital transformation of the travel experience is a priority for the airline industry, in part because of changes outside the industry. We compete with other airlines, of course, but also with the major internet companies that have raised the bar for customer expectations. At the Lufthansa Group, where I’m responsible for the data generated by […]

Continue reading