Featured

Finding patterns in the world around us

Ted Bohn is a Data Scientist and a Research Scientist in Hydrology at Arizona State University.Updating failedTed Bohn is a Data Scientist and a Research Scientist in Hydrology at Arizona State University.

The world is complex and chaotic. At times it seems downright unpredictable. But just under the surface, patterns lurk everywhere. With enough tenacity, we can follow the patterns to underlying rules, leading us to clarity and deeper understanding. We can even leverage these patterns to make predictions, however uncertain they may be.

The search for patterns has been a common thread through my career, linking seemingly disparate fields such as seismology, fisheries, geography, hydrology, and water resources. At its core lies data science, a set of powerful tools for finding and leveraging relationships among data.

I look forward to sharing the beauty and wonder of these sometimes surprising relationships with you in this blog.

CutoffPredictor: avoiding service interruptions with machine learning

As an Insight Fellow, I recently completed a consulting project for Valor Water Analytics. The business problem is this: utility companies deal with substantial time lags between dispensing a resource (e.g., water) to customers, and receiving payment. Customers whose payments are up to a month late receive a late fee. If more than a month passes after the due date and the customer still hasn’t paid, the utility typically cuts off the customer’s water supply. By that point, the customer could have continued using water for up to 3 months without paying for it. Thus, cutoffs are not only bad for the customer, they represent up to 3 months of lost revenues and lost water for the utility. A utility would therefore benefit if it could predict impending cutoffs and avoid them, for example by working with the customer to agree on a payment plan.

My project goal was to create a tool that could predict impending cutoffs from the data available to a utility: customer metadata (location, customer type, meter size), customer billing history (including which payments were late), and customer usage history. It would query the utility’s database, clean the data and compute features, fit a model, make a prediction (on a daily basis), and display the results in a dashboard.

The finished product is CutoffPredictor. More details to come…