Faster off the mark

When a river unexpectedly overflows its banks somewhere in central Europe, Florian Falger often receives a phone call the very next day. The market manager helped develop Endress+Hauser’s Netilion flood monitoring early-warning system. With input from sensors and artificial intelligence, it is capable of predicting if, when and where a flood might happen. “That way, the responsible authorities can make the right decisions at the right time and take protective measures, for example to prevent damage,” says Falger, an industrial engineer by training. The system is unique for its type – and that is not the only thing setting it apart: there was just a year between the start of the project and market rollout.

By any measure, 12 months is not much time to develop a finished product. For a solution that combines sensors with AI based software, though, such pace is comparable to the speed of light. “That it went so fast is thanks to a new, agile department and our open innovation approach,” says Dr Simon Zühlke, a strategy expert at the Endress+Hauser product center for level and pressure measurement. Open innovation means the company is purposefully opening up to outside partners in the active search for innovation alliances. “We now live in a dynamic world where no one can do it all on their own,” says Zühlke, a process engineer. “Pooling skills and knowledge makes it faster to tap into new fields.”

The partners initially experimented with digitalizing storm water overflow basins using measurements and an algorithm. But then, in 2021, catastrophic flooding hit Germany. “Suddenly, there was a major societal need,” Falger says. “We asked ourselves: ‘Could we use the radar sensor and AI to leverage water level measurement and turn it into a flood warning system for small bodies of water?’” A newly set up internal innovation lab got to work on the idea. “There we operate outside of the standard processes, using agile methods, to work on new products and solutions. We cooperate closely with customers – and we enjoy a lot of freedom,” Falger explains.

The result is an optimal blend of an idea, a business start-up and knowledge. Endress+Hauser provided level measuring devices, soil moisture sensors from its IMKO subsidiary and precipitation gauges to supply the required measurements and a cloud platform. Okeanos brought in hydrological knowledge (the founders had written their dissertations on flood forecasting) along with the AI algorithm. This merges measurements with other data in the cloud and delivers a clear forecast, saving users the task of interpreting the numbers. AI training was in cooperation with pilot customers. The system is already in widespread use. “That gives local communities up to 45 minutes of advance warning in the event of a potential flood,” Falger says. “And we want to make that window bigger still.”