As DARROW enters its final year, we are not only highlighting the project’s technical results, but also the people behind them. Throughout this interview series, DARROW partners reflect on the project’s outcomes, the challenges they faced, and the lessons learned along the way.
In this interview, Ruud Peeters from the Dommel Water Authority in The Netherlands shares his perspective on AI implementation in wastewater operations. He offers insights from the intersection of strategy, innovation, and real-world wastewater treatment operations.
Tell us a bit about yourself and your organization. What role have you played in DARROW?
My name is Ruud Peeters. I have been working at the Dommel Water Authority since 2010. My role as an advisor focuses on strategy, policy, and innovation in the broadest sense – both digital and non‑digital. Besides my involvement in DARROW, I am also the program manager for a mechanistic digital twin project that spans the entire system: from the catchment area to the Eindhoven wastewater treatment plant and into the surface water environment.
In your view, what is the most valuable innovation or tool that DARROW has developed?
I see three key innovations:
- The modular design of the integration platform, which enables all academically developed tools to operate both independently and as part of an integrated whole.
- The hybrid approach to training the AI algorithms. Due to the renovation of the Tilburg WWTP, we could not use historical data for training. As a solution, we used calibrated mechanistic models to generate synthetic training data. This makes it possible to quickly adapt the model to changes, such as adding a tank or adjusting control strategies.
- The method for monitoring and predicting biomass. A simple reduced-order model (ROM) based on key biomass kinetics allows us to predict fundamental plant behavior with minimal effort.
If you had to describe DARROW in one sentence, what would it be?
A process for integrating AI into existing wastewater treatment plants, with the potential to achieve full plant-wide control.
What is something you have contributed to DARROW that you are especially proud of?
I am proud of my role as an intermediary: connecting domain knowledge from WWTP operations and their real needs with the modelling challenges that require technical expertise. Helping to translate between these worlds has been my key contribution.
What was one of the biggest challenges you or your team faced, and how did you overcome it?
Wastewater treatment plants, such as our Energy and Resources Factory in Tilburg, were recently designated as critical infrastructure. As a result, very strict cybersecurity requirements were introduced. It took significant time and effort for our organization to develop an AI-driven testing environment that fully complies with these new requirements.
What advice would you give to a future project team taking on something as ambitious as DARROW?
Choose a wastewater treatment plant that is operating stably, without major renovations or TAG recoding underway (TAG codes are unique asset identifiers used in the control system to address specific sensors, pumps and other equipment). Try to keep all key people involved throughout the entire project.
Also, plan several in-person working weeks during the project. Being physically together accelerates understanding — of each other, of complex processes, and of the technical solutions — especially when team members do not yet know each other well.
What kind of change do you think DARROW can bring to how we manage water and resources in the future?
I see two main applications for the DARROW process:
- Plant-wide control: When all subprocesses are modelled and controlled through a master model, the entire system can operate more harmoniously. We know that slow biological processes perform best when they run as steadily as possible. Unfortunately, this part of the DARROW solutions will not be finished by the end of the project, but it remains very interesting to pursue in the future.
- Support for minimally instrumented plants: Facilities with limited sensors can still benefit greatly from the simple biomass ROM model, which requires only minimal additional instrumentation.
Ruud’s reflections highlight a core strength of DARROW: AI implementation in wastewater operations depends on close interaction between operational reality and advanced modelling. His role as a mediator between technological developments and the practical day-to-day in a WWTP illustrates how meaningful innovation emerges when real needs, domain expertise, and advanced tools are developed together.