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, Janelcy Alferes, a researcher from VITO, shares her perspective on data challenges in wastewater treatment: from improving data availability and quality at wastewater treatment plants to developing practical AI-based tools that can be implemented in real operational environments.
Tell us a bit about yourself and your organization. What role have you played in DARROW?
My name is Janelcy Alferes. I work at VITO, the Flemish Institute for Technological Research, which is based in Flanders, Belgium. Our goal is to accelerate the transition to a more sustainable world. VITO de-risks innovation for businesses through interdisciplinary research, innovative technological solutions and large-scale pilot installations. At VITO, I work as an R&D researcher on the topic of Digital Water, covering the entire data chain, from data to smart decision making. My work focuses especially on the integration of physical technologies (e.g. water and wastewater treatment processes) with digital technologies towards innovative and optimized processes.
Within DARROW, I coordinated the tasks aimed at generating enhanced data. Together with my great colleagues Frie Van Bauwel and Maarten van Loo, we developed several data-driven solutions that focus on increasing the availability and quality of data for wastewater treatment plants (WWTPs). These solutions provide reliable and insightful data to be used in further applications such as real-time control, digital twins, process optimisation and decision making.
In your view, what is the most valuable innovation or tool that DARROW has developed?
What makes DARROW unique is the development of not one but multiple practical data-driven tools. These are all part of the DARROW platform and provide optimised data, control and smart decision making for wastewater treatment plants. These are customisable data-driven AI solutions that support the journey of the wastewater sector to more sustainable and efficient utilities, all through the smart use of data!
DARROW introduces a multilevel architecture with data as its core element. At VITO we coordinated the development of the bottom-layer modules focused on data enrichment and augmentation, that means providing data with the right quality and at the right amount. We have made significant progress in bringing these AI-based approaches into practice, while taking into account practical constraints in full-scale scenarios. This also results in a high replication and scalability potential for similar case studies.
If you had to describe DARROW in one sentence, what would it be?
The power of data: DARROW’s data-driven AI solutions will help transform wastewater utilities into more sustainable and efficient systems.
What is something you have contributed to DARROW that you are especially proud of?
My focus has always been on applied research in the water sector, exploiting the potential of combining water treatment technologies, or the ‘physical side’, with the ‘digital side’. This combination creates value with digital technologies supporting and accelerating the transition to more sustainable water systems. DARROW is a perfect example for this: providing AI-powered solutions for WWTPs. The key to success here lies in the intersection between domain expertise and data-driven AI-based approaches to ensure a reliable implementation in practice. In DARROW, I have been able to contribute my expertise in this regard.
What was one of the biggest challenges you or your team faced, and how did you overcome it?
Data-driven solutions are of course based on data. And not only any data, but data that is representative and sufficient for proper development and validation. As we progressed in the project, we realised that the available data was insufficient and/or not representative of the current situation at the demonstration site case study due to several practical reasons, such as a plant upgrade and cybersecurity constraints. We adapted our approach to overcome this challenge. This included adaptating the development and validation methodology, collecting additional operational and sensor data at the case study and engaging further with the plant’s operators, following an iterative approach.
While this could be seen as a drawback, it actually highlighted the real challenges of bringing data-driven solutions into practice in a complex context like WWTPs. The challenge also prompted us to develop and adopt best practices to bridge that gap. This paves the way for future developments and replication opportunities.
What advice would you give to a future project team taking on something as ambitious as DARROW?
Bringing innovative data-based solutions into practice requires considering a variety of factors.
Ensuring that the right input information is available or can be generated is fundamental. Data-driven models must be developed, calibrated and validated under representative conditions to be robust and reliable. This usually follows a continuous interaction process, where domain expertise and an understanding of the operational context are critical to transforming raw data into valuable and effective solutions.
Furthermore, data security is paramount when integrating these solutions into existing IT/OT infrastructure.
Finally, solutions must be adapted to the level of digitalisation in the case study and to specific operational requirements.
On a less technical level, success also highly depends on effective team collaboration and fostering a culture that values data, involving everyone from researchers and engineers to practitioners and operators.
What kind of change do you think DARROW can bring to how we manage water and resources in the future?
The digitalisation of the water sector is a paradigm shift from traditional operations that are no longer capable of tackling current and future challenges towards a smarter way of working. For water utilities, the focus is shifting from water to becoming Water Resource Recovery Facilities (WRRF), where water, energy and other resources are interlinked and resources are optimised. In this transition, data is a key asset. Data-driven and AI/ML based solutions are becoming key to unlock the data value. The methodology and solutions developed within DARROW contribute to this paradigm shift by providing insights and enhancing decision making to optimise resource recovery, improve autonomy and efficiency in WRRFs, paving the way for further development and exploitation.
Janelcy’s perspective makes clear that addressing data challenges in wastewater treatment is essential for unlocking the full potential of digital solutions. By addressing these challenges directly, the DARROW work at VITO shows how data-driven approaches can move beyond ideal conditions and become applicable in practice. It also reflects a broader shift in the sector, where improving how data is generated, interpreted, and used is key to managing treatment plants as integrated systems for water, energy, and resource recovery.