Data Science for Social Good

For the past year, we’ve been identifying different ways to reduce the number and minimize the impact of water main breaks in the City of Syracuse. You’ve read about a variety of initiatives, including installing sensors, testing and replacing valves, and coordinating on infrastructure projects.


However, last week, we added another initiative to the mix. We started working with Data Science for Social Good, out of the University of Chicago, to develop an early warning system for water infrastructure problems - read more here.

We know our water mains are old, and while age might be a contributor to why the mains break, there are likely other factors that contribute to water main failure - material, weather, soil content, etc. By building a model that considers many different attributes, the Data Science for Social Good team can identify and show which factors are important when considering why water mains break.


Using data can help confirm assumptions about why mains break, and can ensure that contributing factors to breaks are documented.  This will provide the Water Department staff with good, data-driven information, so that they can appropriately plan for future water projects.


The Eric & Wendy Schmidt Data Science for Social Good Fellowship is a summer program that trains aspiring data scientists to use a variety of techniques to complete projects that have a positive social impact. Other projects have included building tools to proactively reduce blight in neighborhoods, creating early intervention systems to reduce adverse police interactions, and more. You can see the full list here.


We’ll be working with the Data Science team throughout the summer, with the end goal of getting a list of the riskiest water mains in the city. From there, we’ll use the Data Science for Social Good information to build out a plan for water main replacement. Ultimately, we hope that this can help the Water Department proactively replace mains rather than constantly respond to breaks.


There will be more updates on this project here over the next couple of months, and we should have some final reports by the end of August.


If you have thoughts about why a water main breaks, leave it in the comments section below!