City Of Topeka, Kansas: Predict, Prioritize, Protect Case Study

Three engineers. $6.5 million in waterline replacement every year. Hundreds of miles of aging pipe. For Topeka's utility, age and break history were no longer enough to build a capital plan that could be defended to the dollar. Here is how the team paired likelihood of failure with consequence of failure, using a machine learning model trained on data from more than 200 utilities, to decide where capital should go first. Braxton Copley, Director of Utilities, speaks to managing a replacement program with a very small staff, and the results cover model accuracy across two years and the savings measured per capital project. Explore the numbers behind Topeka's approach.