ExplAIn brought together 15 partners from Sweden, Germany, and the Netherlands. Södra, alongside companies such as ABB, Viking Analytics, and academic institutions, contributed to making AI more transparent, reliable, and user-friendly for operators and engineers in industrial settings.

Södra’s focus has been on predictive maintenance and the ability to anticipate developments in its processes.

The developed AI solution can be used to predict what will happen in the cooking process – a crucial step for improving stability and pulp quality. By using AI to analyze process data and present it through methods developed within ExplAIn, earlier warnings and deeper insights into the process can be achieved. This enables action before problems arise.

The project has also demonstrated how explainable AI – known as XAI – can foster trust and robustness in industrial environments. By placing humans at the center – both during development and usage – Södra has gained access to new methods and tools that strengthen the future use of AI in the pulp industry.

“Through ExplAIn, Södra has gained further opportunities to apply industrial AI. The project has clarified the importance of putting people at the center to create the best possible conditions for trust and robustness. It has provided us with valuable methods and tools for the future,” says Andreas Darnell, Senior Development Engineer OT – Södra Cell Technology Development.

Image caption: From left: Andreas Darnell, Hugo Wärnberg, UX Researcher & Designer ABB, and Yanqing Zhang, Senior UX Researcher & Designer ABB.