University of Bergen's Professor Ivar Rønnestad and researcher Sergei Budaev, who are developing FishMet in collaboration with Vestlandets Innovasjonsselskap (VIS).
Photo: University of Bergen
A research team in Norway is working on a digital twin model that could change how salmon and trout are fed.
The project, called FishMet, is being developed at the University of Bergen by Professor Ivar Rønnestad and researcher Sergei Budaev, in collaboration with technology transfer organization Vestlandets Innovasjonsselskap (VIS).
The developers describe FishMet as a proof-of-concept designed to support precision feeding in aquaculture, and claim their system could help farmers reduce feed waste, improve fish welfare and boost growth rates.
According to a news release from the University of Bergen, FishMet is built around a digital twin framework that integrates biological and environmental information into a “virtual fish”. The system simulates appetite, digestion, metabolism and growth by drawing on physiological feedback loops, rather than relying solely on opaque machine learning models.
It can incorporate inputs such as fish size, feed type, feeding frequency, water temperature, oxygen levels and behaviour. The output includes estimates of feed intake, growth efficiency and gut transit times, as well as indicators of stress or motivation. The model is modular and can simulate single fish or entire populations, the researcher say.
“We aim to create a transparent digital salmon that combines AI with decades of biological knowledge, serving both as a research tool and a practical aquaculture predictor, especially in situations lacking data,” said researcher Sergei Budaev.
The concept remains at an early stage, but pilot studies have produced encouraging results in rainbow trout and Atlantic salmon, including predictions of gut transit times and growth performance, the research team said.
While further validation is needed, the researchers argue that FishMet could provide an alternative to “black-box” AI systems by offering a more transparent, biology-based approach. They suggest it may help farmers lower feed costs, improve production efficiency and reduce environmental impacts.
Looking ahead, the team says FishMet could be expanded to other species or life stages such as smoltification and maturation. VIS has now made FishMet available for exploratory licensing opportunities, which could allow industry partners to test potential applications.