Work ยท Applied Analytics
Predictive Water Infrastructure
Using historical, operational, and infrastructure data to make a difficult public-works decision more proactive.
The problem
Water-main failures are expensive, disruptive, and difficult to anticipate. The city needed a better way to prioritize a large, aging network before emergencies forced the decision.
What we tried
We worked with research partners to combine the records and local knowledge that could help identify risk, then translated the resulting model into a practical planning tool.
My role
I helped define the operational problem, connect the people who understood the infrastructure with the people building the model, and make sure the output could support a real decision.
What happened
The work created a defensible way to prioritize risk and strengthened the case for more proactive infrastructure investment. It also became a useful example of research meeting a real city problem.
What I learned
The model is rarely the product. The product is a better decision, made by people who understand why the analysis belongs in the room.