Advancing smart aquaculture: Cost-efficient strategies for climbing perch cultivation using AI-based models

Abstract

This study introduces a hybrid AI-based optimization framework to enhance climbing perch aquaculture in smart farming systems, targeting improvements in both productivity and cost-efficiency. By integrating Taguchi experimental design with reinforcement learning and metaheuristic algorithms, the model identifies critical environmental and operational parameters that influence fish growth and economic outcomes. A multi-objective regression approach is applied to optimize key factors including water temperature, dissolved oxygen, pH, feeding schedules, and stocking densities. The RL-AMIS method, which combines Pareto front analysis and TOPSIS, successfully balances trade-offs between cost and productivity. Experimental results show a 15.3 % reduction in operational costs and a 17.8 % improvement in growth efficiency compared to conventional methods. These findings demonstrate that AI-driven optimization can enhance scalability and adaptability in aquaculture systems, supporting sustainable production strategies.