Carollo Engineers and partners have released a free guidebook to help water utilities evaluate, deploy and maintain artificial intelligence and machine learning tools in potable reuse systems.
The AI & Machine Learning Guidebook for Potable Reuse was developed through research funded by the U.S. Bureau of Reclamation’s Desalination and Water Purification Research programme.
The guidebook focuses on potable reuse, but Carollo said many of the underlying principles also apply to conventional drinking water and wastewater treatment systems.
It was developed in collaboration with Carollo Engineers, Yokogawa, the National Water Research Institute and Baylor University.
Practical guidance for AI adoption
The guidebook covers the full AI and machine learning implementation lifecycle, from foundational concepts and data management through to model development, pilot testing, cybersecurity, workforce readiness and long-term maintenance.
It also sets out real-world applications including process optimisation, predictive maintenance, water quality forecasting, digital twins and fault detection.
“Artificial intelligence and machine learning have the potential to help utilities make better use of the vast amount of operational data generated every day,” said Andy Salveson, vice president at Carollo Engineers and principal investigator for the project.
“This guidebook provides a practical path forward for utilities interested in exploring these technologies, from understanding the fundamentals to successfully implementing and maintaining tools that support more informed operational decisions.”
Potable reuse and operational resilience
Potable reuse systems generate large volumes of process and water quality data. AI and machine learning tools can help utilities identify abnormal performance, forecast water quality, optimise process control and support operator decision-making.
The research project behind the guidebook is titled Data-Driven Fault Detection and Process Control for Potable Reuse with Reverse Osmosis and Membrane Bioreactors.
Carollo said the guidebook was shaped by input from a Utility Advisory Board and an Independent Expert Panel representing utilities, researchers and consultants.
The publication emphasises that AI adoption is not only a technical challenge. Data quality, governance, operator engagement, cybersecurity and organisational readiness are all presented as essential to successful implementation.
Phased approach recommended
The guidebook recommends a phased implementation approach, allowing utilities to build confidence, validate model performance and reduce risk before wider deployment.
This includes identifying high-value use cases, assessing available data, testing models in controlled environments and setting clear responsibilities for long-term monitoring and maintenance.
Carollo said the resource is intended for utilities, researchers, regulators and water professionals looking to understand how AI and machine learning can be applied responsibly in potable reuse and related treatment applications.
The guidebook is available to download from Carollo’s website.
FAQs
What is the AI & Machine Learning Guidebook for Potable Reuse?
It is a free guidebook designed to help utilities evaluate, implement and maintain AI and machine learning tools in potable reuse systems.
Who developed the guidebook?
The guidebook was developed by Carollo Engineers, Yokogawa, the National Water Research Institute and Baylor University through U.S. Bureau of Reclamation-funded research.
What topics does the guidebook cover?
It covers data management, model development, pilot testing, cybersecurity, workforce readiness, long-term maintenance and use cases such as predictive maintenance, water quality forecasting and digital twins.
Is the guidebook only for potable reuse?
Potable reuse is the main focus, but many of the principles also apply to conventional drinking water and wastewater treatment systems.
