Creating a low-cost, real-time water quality monitor using AI
The U.S. is facing a drinking water crisis.
77 million Americans get their drinking water from a system that reported violations of the Safe Drinking Water Standards. As a result, many households need to trust that their tap water is safe to drink. The result is that U.S. households spend over $44 billion each year on bottled water and water filters.
Low-income and minority communities are most likely to face an endemic drinking water crisis. Because they do not trust their tap water, these households spend a larger percentage of their income on tap water alternatives such as sugary drinks, which lead to poor health outcomes.
Working with UCLA, we set out to solve a critical technology gap.
There are expensive and complex systems for industrial users and inexpensive but unreliable solutions for consumers. But, there isn't a reliable, affordable, and simple solution that households and communities can use to monitor their water - one that provides warnings and guidance on how to use the data.
Working with UCLA's School of Civil and Environmental Engineering, we set out to develop the first low-cost, real-time tap water quality monitor. The system measures the five most important water quality parameters and uses AI/ML models to determine the water's health and risk factors
"If we had this technology in 2014, the Flint Water Crisis might not have happened" - Dr. Eric MV Hoek, UCLA
We believe that by overlaying tap water data with other public health datasets, we can better understand the relationship between water quality and health outcomes in at-risk communities.
