Follow Datanami:
July 1, 2020

OmniSci Powers New Website Enabling Public to View House-by-House Information On Flint Water Crisis

SAN FRANCISCO, Calif., July 1, 2020 – OmniSci, provider of accelerated analytics, working in close partnership with water infrastructure analytics consulting company BlueConduit, announced the debut of Flint Service Line Map (www.flintpipemap.org), a public website that maps up-to-date information about residential water service line replacements in the city of Flint, Michigan. These water service lines are the pipes that deliver each home their water. If the pipes are made of lead, they can contaminate that home’s water with lead. The problem: Flint, like most other cities, did not know exactly which pipes were lead. Presented in house-by-house detail, the map allows residents to easily find out about their known or likely water service line material, along with repair dates and other useful information.

Flint Service Line Map is the first site to utilize artificial intelligence and machine learning to predict, with high accuracy, the incidence of lead pipes in residential service water lines. The site is provided as a public service and shows the same assessment and status information used by contractors in the city’s FAST Start pipe replacement program. Residential properties are color-coded individually, according to likelihood for having lead service pipes. Clicking on a location reveals street address, pipe verification date, repair updates and links to pertinent city information. BlueConduit collaborated with NRDC (Natural Resources Defense Council), one of the country’s leading environmental organizations, to ensure the map adhered to best practices in public health communication.

OmniSci’s innovative accelerated analytics computing platform powers the interactive experience developed by BlueConduit, giving residents the power to explore the predicted likelihood of lead or galvanized steel at each residence. A breakthrough in data analytics, the OmniSci platform utilizes GPUs in a parallel processing architecture to analyze billions of rows of geospatial data in milliseconds.

“As we addressed the incredible complexity of predicting the presence of lead pipes in Flint, our model incorporated a massive amount of data from dozens of sources. We were extremely pleased with the power of the OmniSci platform to integrate and process this information,” said Jared Webb, BlueConduit’s chief data scientist. “We looked at other technologies, but OmniSci was the only solution that was able to interactively query, filter, and render the geospatial data we had.

Beginning in 2016 in response to its water crisis, the City of Flint has been replacing lead or galvanized steel service lines to residential homes under its FAST Start program. Due in large part to the predictive algorithm developed by BlueConduit and enabled by OmniSci’s accelerated analytics technology, the hit rate for finding lead or galvanized service pipes in Flint increased from as low as 15% without the use of the algorithm, to 80% with the algorithm.

“The geospatial nature of the Flint problem, combined with the size and scope of the analytics challenge, was an ideal application for the OmniSci platform,” stated Aaron Williams, Chief Advocate for OmniSci. “It’s been immensely gratifying to see our analytics technology at work through the OmniSci for Good program, increasing the speed in which lead has been identified and removed from the Flint water system.”

Through 2019, the City of Flint has checked service lines at 24,304 residences and replaced 9,448 lead or galvanized steel lines. It predicts more than 1,000 homes still require replacements and anticipates completing the project by the end of 2020.

“The community, environmental health, and economic benefits of this project can’t be overemphasized,” stated Eric Schwartz, co-founder of BlueConduit and professor at the University of Michigan. “Helping the Flint community switch quickly and most efficiently to safer water infrastructure is an ideal example of how big data analytics can be used to solve complex problems with a positive social impact. BlueConduit has worked to help stretch taxpayer dollars farther so that they can remove lead service lines from more residents’ homes and do so quicker. BlueConduit is proud to have welcomed OmniSci as a key partner in our initiatives for addressing these infrastructure and informational challenges that continue from the Flint water crisis.”

About OmniSci

OmniSci is the pioneer in accelerated analytics. The OmniSci platform is used in business and government to find insights in data beyond the limits of mainstream analytics tools. Harnessing the massive parallelism of modern CPU and GPU hardware, the platform is available in the cloud and on-premise. OmniSci originated from research at Harvard and MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). OmniSci is funded by GV, In-Q-Tel, New Enterprise Associates (NEA), NVIDIA, Tiger Global Management, Vanedge Capital and Verizon Ventures. The company is headquartered in San Francisco. Learn more about OmniSci at www.omnisci.com.

About BlueConduit

BlueConduit leverages machine learning to ensure that residents of every community regardless of means have access to clean water.  Growing out of research at the University of Michigan starting in 2016, BlueConduit has used predictive models to help cities identify which homes have lead service lines, providing a home-by-home inventory to guide their removal in a cost effective way.  Their work in the field has appeared in The New York Times, The Chicago Tribune, The Atlantic, and the Smart Water Networks Conference. The team’s research was recognized as one of the top four award-winning papers at the Knowledge Discovery and Data Mining Conference, the top-ranked data science conference. BlueConduit is working with a growing list of utilities and municipalities throughout the midwestern and eastern United States. BlueConduit’s headquarters is in Ann Arbor, MI. Learn more about BlueConduit at www.blueconduit.com.


Source: OmniSci

Datanami