Fusion of transformer-based deep learning and Monte-Carlo fish growth simulation for aquaculture smart transformation

Purpose Develop a deep learning-based fish growth model to improve the accuracy of fish growth predictions and optimize feeding strategies in open-sea aquaculture cages. Design/methodology/approach We employed the Monte Carlo approach to generate big data for training transformer-based deep learning models to predict fish growth trajectories during cultivation. In generating big data, each key factor of the fish growth model is modeled with a probability distribution parametrized by real-world fish growth data from IoT-based monitoring systems and open weather datasets. Findings Minimal prediction errors from 2.02 to 3.01% for weight; growth rate errors consistently below 2.5% and feeding amount and the meat conversion rate exhibit slightly higher but still acceptable error margins (∼7.4–8.0 and ∼5.1–5.4%, respectively). Research limitations/implications Dependency on accurate initial parametrization of the probability distributions and the reliability of data collected from IoT systems or other dataset sources. The model should be robust against varying environmental conditions, and it has limitations in application to different types of aquaculture environments. Practical implications First, the techniques can enhance precision in aquaculture by providing accurate fish growth predictions and optimized feeding strategies. Second, reducing feed consumption not only lowers production costs but also minimizes the environmental impact of excessive feed waste. Lastly, the use of IoT-based monitoring systems and smart feeding machines can streamline aquaculture operations, making them more efficient and sustainable. Social implications Optimizing feeding strategies and reducing feed waste promotes more sustainable aquaculture practices, which can lead to a reduction in the environmental footprint of fish farming and benefit ecosystems and biodiversity. Enhanced precision in fish growth prediction and optimized feeding can lead to more efficient production of fish, contributing to food security by ensuring a stable and reliable supply of high-quality fish protein. Lower production costs and improved efficiency can make aquaculture more profitable, potentially benefiting local communities economically, especially those dependent on aquaculture for their livelihoods. Originality/value The study uniquely combines deep learning with the Monte Carlo approach to overcome the challenge of obtaining high-quality big data for fish growth prediction. This innovative combination allows for the generation of robust training datasets from limited real-world data. The use of a transformer-based deep learning model to predict fish growth trajectories is a novel application in the field of aquaculture. Also, the transformer’s application to aquaculture demonstrates a creative adaptation of advanced machine learning techniques.