Director of Strategic Initiatives , Global Water Challenge / Global Environment & Technology Foundation
AI and Augmented Humanity
Founded in 2006, Global Water Challenge (a subsidiary of the Global Environment & Technology Foundation) is a coalition of leading organizations committed to achieving universal access to safe drinking water, sanitation and hygiene (WASH). With leading companies, civil society partners, and governments, GWC accelerates the delivery of safe water and sanitation through partnerships that catalyze financial support and drive innovation for sustainable solutions. Through GWC's innovative public-private partnerships, over 1 million people have been reached with clean water access.
Water is so massively important in improving so many issues around the world -- gender equality, education, public health, economic development - and delivering safe and sustainable water to communities around the world, especially in sub-Saharan Africa, can be a simple process -- drill a hole in the ground, get a water point, get access to fresh water. Unfortunately, these water points often break within 4 years of being installed, and when that happens, local populations who don't have the tools and skills to fix it are forced to go back to old water sources that are unsafe or inconvenient. Being able to do predictive maintenance on wells while figuring out where to best install wells that will serve the broadest populations could have a massive impact on delivering safe and sustainable water.
Brian Banks and GWC helped create the Water Point Data Exchange (WPDx), a global framework for openly sharing water point data that can be easily shared, accessed, and used to drive evidence-based policy and funding decisions. With over 500,000 water point records across more than 50 countries, the organization had access to a ton of potentially valuable data...but no real way to make sense of it, and no data science resources to execute on it. Around 2017, Brian met Chandler McCann, a data scientist at DataRobot who had worked on the WPDx as part of his grad school capstone project. Using DataRobot's automated machine learning platform and Chandler's data science skills, they were able to build predictive models using the WPDx data that could accurately predict which water points were most likely to break and when, allowing the Water Ministry of Sierra Leone to take preventative measures through predictive maintenance.
Following a pilot in two regions, Sierra Leone is now requiring the use of these tools across the entire country when planning for water services. This directly impacts water services at more than 26,000 water points that serve more than 2,000,000 people on a daily basis. They were also able to select locations for water points not by using assumptions (as was previously done) but by predictive modeling. District planning officers are using the prediction tools -- combined efforts between DataRobot and GWC -- for budget and location decisions.
- Predictive models used to make water point decisions on new installation + predictive maintance for more than 26,000 water points, serving more than 2,000,000 people on a daily basis in Sierra Leone
- Water Point Data Exchange (WPDx), a global framework for openly sharing water point data that includes access to over 500,000 water point records across more than 50 countries
- DataRobot, an automated machine learning platform that ingests data to build highly accurate predictive machine learning models without requiring the need of a data scientist
- a web-based platform that combines the two, allowing users to filter by location and other factors to get water point predictions
Brian didn't have a background in data science, and GWC didn't have the resources they needed. The disruptive technology of automated machine learning platform + the diligence of Brian in collecting and maintaining the records in WPDx bridged that data science gap. "Working with DataRobot has systematically solved every single one of the challenges we faced in delivering solutions to governments," said Brian.
"With Brian, we're able to quickly drag and drop this dataset in to DataRobot, and his reaction was ecstatic. We just accomplished in one hour what he'd been trying to do for the last two years," said Chandler.
Brian and Chandler went over to Sierra Leone to present at the Ministry of Water to share the tools and teach what is AI and machine learning and how and why they can trust these predictions and data. They held a session for more than 60 people from all regions and levels of water management in the government and taught them how to use the tools and interpret them, as well as why the data that locals were collecting in the field was important.
About Your Organization
Founded in 2006, Global Water Challenge (GWC) is a coalition of leading organizations committed to achieving universal access to safe drinking water, sanitation and hygiene. With leading companies, civil society partners and governments, GWC accelerates the delivery of safe water and sanitation through partnerships that catalyze financial support and drive innovation for sustainable solutions.