Sustainable industrial production of fish using AI and advanced data analyses

Purpose and goal:

The objective of the study was to make industrial land-based production of fish more circular, sustainable, and profitable using AI and advanced data analysis. Within the framework of this preparatory feasibility study, we have made this possible by implementing and adapting several different data-driven methods for optimizing land-based fish production.

Expected results and effects:

Outcome objectives included evaluating state-of-the-art questions for use in data-driven, adaptive production of fish in the RAS. We have implemented and developed methods for acoustic and optical identification of behavior and activity in connection with feeding, as well as implemented and further developed a method for estimating biomass with underwater sound.

Approach and implementation:

The project was designed with clear work packages and a clear division of responsibilities. It was a challenge to get so many different things done in such a short time, but to have this opportunity to start our work towards more sustainable and profitable fish farms on land was very valuable.