Improving Freshwater Use with PI + AI through CONNECT

This month’s guest blog by Dan McNulty explores the origins of ITI Group’s entry in the 2026 AVEVA Partner Games, which will compete for gold at the finals in Milan. Dan explores how publicly available data and AI can be used to help farmers and water users make better decisions without adding more sensors or infrastructure. 

Every time you eat at the dinner table, there is a global problem which has been overcome on your behalf – a problem you share with almost every plant and animal on earth – and many segments of industry, too. 

Some global challenges remain unsolved simply because they don’t align with traditional business goals – there are no ‘shareholders’ or ‘KPIs’ to worry about – yet they affect all of us, every single day. 

Data science and AI solutions tend to pop up wherever there are profits at stake: process optimisation; anomaly detection; OEE. Modern industrial sites are full of niche solutions for enhancing various isolated processes. However, AI is better geared for true ‘big data’ problems – issues larger than the walls of closed systems: global problems, those too complex and involved for solely human approaches to overcome. 

The Data Gathering Challenge 

In responding to AVEVA’s call to ‘Build solutions for real world customer challenges’ in the 2026 AVEVA Partner Games, we have chosen to face one of the most pressing universal problems – one which receives surprisingly little analytical attention. 

In 2024 my team tackled wind turbine failures using ML, but without data on real failures, read from real instruments. Without a paying customer, the data required to solve problems is surprisingly hard to acquire! We tried for the pinnacle of maintenance: predicting failures pro-actively… but ended up with far too little data to work with. So this time, we’re leveraging open data sources. 

So rather than the pinnacle – where data is hard to find – we chose to start low. Very low. In fact, at the base of all human needs, a universal challenge we share with even plants and animals: shortages of usable fresh water.

Ecology and industry are often intertwined – if we can improve water availability of water sources, we can strengthen and protect every downstream industry, including key water users in manufacturing, mining and other industrial sectors.  

The Water Management Challenge 

A family member inspired me to investigate open access to UK water and hydrology data, and the problem crystallised at once: Water Scarcity. 

In the UK, water seems generally plentiful from day to day… until the weather turns, water networks fail, and the gap between supply and demand suddenly widens. As it turns out, our water situation is more fragile than we dare to dream.  

For millennia, humans have struggled to keep water resources in check when the weather is dry and hot and demand is great.  

As it turns out, the modern scientific community still takes this problem seriously, and have created some amazing solutions here and there. Impressive projects like Kibbutz Hatzerim use intensive irrigation and water reclamation in hot, dry, otherwise unfarmable regions.

Today, water scarcity modelling is all about understanding networks of water systems, and simulating various scenarios which put strain on these networks. It can then take decades to create robust strategies at the national or regional level, followed by a very patchy uptake and unpredictable results. Many solutions require land operators to opt-in to gather and provide even more data. All of this takes a great deal of time. 

The hope is that building on the research, AI assisted models can help tweak real human behaviour much more quickly. In the future, AI tools will continuously adapt models to emerging data and seasonal observations – but that’s still a bit over the horizon. 

If we could predict water source behaviour for even a few days out of a month, we can help water users find better water usage strategies without putting crops at risk.  

Of course, you could just look out the window and hope for the best – but modern weather forecast models analyse hidden factors like soil moisture, and together with long range data offers the potential to turn intangibles into actionable insights.

The Farmers’ Challenge 

Our team of crack engineers at ITI sought out practical advice from subject matter experts: the knowledgeable folks at the Hutton Institute together with Scottish Environmental Protection Agency (SEPA) records helped us figure out where to start. As it turns out, most water is wasted in irrigation, so this was our best target.  

From the top view, the big problem is simple enough: Farmers care about crops. Crops depend on soil. Soil is fed water by rainfall and irrigation. Irrigation mechanisms are fed by rivers. Rivers are fed by rainfall. Rivers exist in catchments. And so by modelling these steps we can try to connect the top-level problem with local datasets. 

This problem is close to home – there’s soil at risk from drought all over the east coast of Scotland. Fortunately, there’s plenty of publicly available weather and water resource data available. All we need is to gather it, process it using the right tools, and create useful feedback mechanisms that would help farmers trying to care for their crops. 

As the biggest freshwater users, farmers already face pressures on all sides to grow crops profitably: use less water and risk losing the crop, or use more and risk shortages? National water management is tricky, and usually met with mixed responses. Our aim is to build a community-oriented solution to lighten the burden, using readily available data. 

The Milan 2026 Challenge 

As we head to the finals in Milan, our team Cache Crops will push this approach and see how far we can take it, using AI tools to predict water resource depletion, drought intensity and resource recovery to aid users across the land. At this stage it’s difficult to say which of these will be the most useful or effective, and so we still have some distance to travel before arriving at our final solution. 

Coming to AVEVA World 2026? Get in touch using our contact form below if you’d like some time with Dan and the team to take a closer look at their final entry.