Multi-objective configuration of water quality indicators in tilapia production systems using artificial intelligence

Resumen

Objective: To constrain the operational costs of water quality management within acceptable thresholds while preserving “excellent” quality standards, thereby ensuring optimal growth conditions for tilapia.

Design/methodology/approach: Aquaculture has emerged as a key global source of protein. The application of Artificial Intelligence (AI) to manage water quality indicators enables optimization of efficiency, sustainability, and costs. However, the multi-objective integration of parameters and their joint application in aquaculture systems remains an emerging challenge.

Results: The optimization results provide practical references for the design of decision-support systems in aquaculture plants, offering a structured methodology to balance water quality improvement with cost minimization.

Limitations on study/implications: The study was based on simulated data validated through the literature. The current value of the research lies in its conceptual and methodological foundation. Future work could integrate real-time data obtained from sensors.

Findings/conclusions: The proposed methodology demonstrates the potential of combining Artificial Neural Networks (ANNs) and Multi-Objective Genetic Algorithms (MOGA) as a decision-support tool in aquaculture management. By optimizing both water quality and costs, it contributes to improving the efficiency, resilience, and sustainability of intensive tilapia farming.