August 7, 2020 Digitalisation

AI Solutions for the Water Sector: Interview with Prof. Zoran Kapelan

Artificial Intelligence (AI) is a key technology which can make a positive difference in our world. In the journey towards the digitalisation of the water sector, AI plays an important role to ensure a more reliable and secure water supply.

The IWA Digital Water Programme has published in August 2020 the white paper  ‘Artificial Intelligence Solutions for the Water Sector‘. Authored by Professor Zoran Kapelan (Delft University of Technology), Emma Weisbord (Digital Water Consultancy) and Professor Vladan Babovic (National University Singapore), the white paper presents examples that highlight and examine the topic of AI and its applications in the water industry. They are selected to demonstrate that AI-based solutions can address real challenges and provide tangible benefits to the water sector.

On the occasion of the launch of the publication, the International Water Association has interviewed Professor Zoran Kapelan who gives insights on the current and future roles of AI in and for the water sector.  The IWA Fellow Professor Kapelan is Chair and Professor of Urban Water Infrastructure in the Department of Water Management at Delft University of Technology.


IWA: Today, we will talk about Artificial Intelligence or AI. So, Professor Kapelan, could you please tell us how AI is used in the water sector?

Prof Kapelan: AI nowadays is used in a number of different ways. But before going into details, let me just say what do I mean by AI: this is a widely used term and different people mean slightly different things by it. Personally, by AI, I mean a wide range of computer systems and technologies that can ideally perform tasks that would normally require some form of human intelligence. So, AI refers to systems and technologies that make use of tools like machine learning, computer vision, data mining and so on. AI has already made its way in the water sector and its use is actually increasing as we speak. So, let me give you some examples on how AI can be used.

AI has been used in various early warning systems such as computer systems that can detect leaks and discolouration events in water distribution systems or blockages and collapses in wastewater systems; also different failure events at treatment works. AI has also been used for improve asset management, for example, for condition assessment of pipes that make use of computer vision and machine learning techniques. It has been used to predict the deterioration rate of pipes, as well as for various forecasting purposes, e.g. rainfall, water treatment plant inflow, sea currents or water consumption. It has also been used for controlling wastewater plant, scheduling pumps, and, last but not least, for various risk assessments. You can read more about these and other examples of AI solutions in the white paper.

IWA: Thank you for your insights and explaining the use of AI. Based on your experiences, how do you think AI has improved water services for both the operators and the customers?

Prof Kapelan: Another good question, thank you for that. The AI has actually improved the water services in different ways. Rather than just listing all of these, let me give you an example of the benefits that were achieved from the AI-based events detection system that features in the white paper. This is something I developed together with a PhD student in the UK a while ago: the system takes data from pressure and flow sensors in the field, it processes this data and raises alarms when it detects events like pipe bursts or equipment failures. This has resulted in multiple benefits for the water company: it has reduced operational costs for dealing with bursts but also in financial rewards provided by the industry regulator for reducing the customer supply minutes lost. In addition, this system reduced leakage which resulted in reduced energy for pumping and lower carbon footprint. All of this has a consequent clear and positive effect not just for the company bus also for the environment. The main benefit, of course, goes to the customers who are provided improved water service by the company (which, in this case, is 7 million people and 200 000 business customers), via reduced interruptions to water supply.

IWA: Thank you for explaining this; it is clear that AI is improving the way water is managed and delivered, so what do you see as the future of AI in the water sector?

Prof Kapelan: This is a really important question. I think AI has an increasingly important role in the water sector and it is likely to increase in the future. We are currently seeing many established companies using AI solutions as well as many start-ups (which nowadays are popping out like mushrooms after the rain!). I can see water utilities investing further into increased observability of their complex systems by introducing additional sensors and I can see this resulting in even a larger amount of data that will need to be collected and processed and it is only natural to consider AI in this context. I can also see water utilities investing more in the so called actuators, various devices, like pumps and valves that can be remotely controlled and can help further automate the operation. I can also see AI being combined with other related technologies like virtual reality (VR) or Internet of Things (IoT), which is actually already taking place. So, as you can see, many current AI based technologies are likely to mature further in the future and, as a consequence, I am expecting wider applications in the water sector.

What I do not see, at least not in the near future, is AI-based solutions replacing people. I personally would like to see AI providing additional support to human operators thus enabling them to do their job better based on more objective information extracted from large amount of data. Based on all of this, I am quite optimistic and think that the future of AI in the water sector is bright!

IWA: Many thanks for that. So this is the end of our interview. Thank you Professor Kapelan for joining us and thank you to all our readers. If you would like to know more about Digital Water please join the IWA Digital Water Group on IWA Connect. Thank you!