Introduction
Economic success in recirculating aquaculture systems (RAS) is not guaranteed. High CAPEX and substantial OPEX for feed, energy, labour and system operation create narrow margins for operational error. Small deviations in feeding efficiency, mortality, or system performance can quickly translate into significant financial losses (Martins et al., 2010; Badiola et al., 2012).
At the same time, RAS generates increasing volumes of operational and biological data. However, value is only created when this data is translated into actionable decisions with measurable economic impact (F��re et al., 2018). In particular, fish behaviour, provides an integrated indicator of feeding response, stress and health (Martins et al., 2012).
The objective of this study is to quantify the economic value of AI-driven operations and maintenance (O&M) support in RAS by defining and evaluating key use cases in RAS and to prioritise their implementation based on financial relevance and technical feasibility.
Material and Methods
Within the Mariculture 2 project at the Rostock Ocean Technology Center, a use-case valuation was developed based on a marine RAS investor business (≈321 T production; farm gate price ~14.5 ���/kg).
Four operational categories were analysed: (1) technical system operation; (2) water quality and biological treatment; (3) fish behaviour and feeding; (4) operational competence and knowledge transfer.
Within each category, 3–4 use cases were evaluated against their effect on cost drivers (feed, labour, energy, oxygen, chemicals, CAPEX allocation), and key financial drivers (mortality, feed efficiency, labour effort and system downtime). Measurable signals were linked to operational interventions and translated into annual economic impact (��� per production unit).
Results
Total economic potential for the application of AI and machine learning based O&M support was estimated at ~1.3 M���/year per production unit (Fig. 1).
Fish behaviour and feeding showed the highest impact (~482 k���/year), driven by feeding optimisation and early detection of stress and health issues. Water quality management contributed ~324 k���/year through stabilised production conditions. Competence and knowledge transfer reached ~309 k���/year via improved decision-making and standardisation. Technical system optimisation showed lower value (~202 k���/year) but higher implementation readiness.
Fig. 1., Economical potential of the four use-case categories evaluated.
Discussion
Highest value is generated where data directly focus on the fish as object of interest. Fish behaviour and specifically group behaviour integrates feeding, water quality and health, making it a key leverage point but also technically demanding.
To address group behaviour evaluation, a pre-project will validate camera-based behavioural monitoring of fish groups in large-scale RAS environments. In parallel, an implementation roadmap towards a minimum viable product (MVP) is developed, defining IT/OT architecture, data integration and decision-support tools to roll out RAS O&M support.
All activities targeting towards the implementation of the AI driven O&M support to be implemented with the first floating RAS, planned to be operational in 2027.
References
Badiola, M., Mendiola, D., Bostock, J. (2012). Recirculating aquaculture systems (RAS) analysis: Main issues on management and future challenges. Aquacultural Engineering. 51, 26–35. https://doi.org/10.1016/j.aquaeng.2012.07.003
F��re, M., Frank, K., Norton, T., Svendsen, E., Alfredsen, J.A., Dempster, T., Eguiraun, H., Watson, W., Stahl, A., Sunde, L.M., Schellewald, C. (2018). Precision fish farming: A new framework to improve production in aquaculture. Biosystems Engineering. 173, 176–193. https://doi.org/10.1016/j.biosystemseng.2017.10.014
Martins, C.I.M., Eding, E.H., Verdegem, M.C.J., Heinsbroek, L.T.N., Schneider, O., Blancheton, J.P., d'Orbcastel, E.R., Verreth, J.A.J. (2010). New developments in recirculating aquaculture systems in Europe: A perspective on environmental sustainability. Aquacultural Engineering. 43(3), 83–93. https://doi.org/10.1016/j.aquaeng.2010.09.002
Martins, C.I.M., Galhardo, L., Noble, C., Damsg��rd, B., Spedicato, M.T., Zupa, W., Beauchaud, M., Kulczykowska, E., Massabuau, J.C., Carter, T., Planellas, S.R., Kristiansen, T. (2012). Behavioural indicators of welfare in farmed fish. Fish Physiology and Biochemistry. 38(1), 17–41. https://doi.org/10.1007/s10695-011-9518-8