Summer 2022 –  A data-driven approach for water safety plans in sustainable buildings to predict and prevent disease 

Project Director: Hamilton

The team conducted a comprehensive field study to target the source of Legionella contamination in a green building.  This work uncovered the water softener and an expansion tank as an understudied source of contamination.   The study will be detailed in a publication in Frontiers in Water to appear in early 2023.

In addition to Legionella sampling, the project team collected sensor and other water quality data.  Machine learning (ML) algorithms are being developed to predict chlorine residual; the goal is that such predictions could inform facilities management software tools. Initial results are encouraging. The figure below shows reasonably good ML model predictions for chlorine residual over a 24 hour period.

predictive residual chlorine models chart