Architecture Design for the Implementation of a Water Quality Prediction System in Aquaculture Systems with Big Data

The success of aquaculture production relies heavily on the effective monitoring and control of various variables throughout the cultivation process. Traditional data collection and processing methods fall short when handling large volumes of data. Therefore, integrating artificial intelligence (AI) and big data techniques is essential. This study aims to design and evaluate architecture for aquaculture's water quality prediction system, leveraging Big Data to enhance fish farming management. The proposed architecture was developed using deductive and inductive methods and analyzed through synthetic analytical techniques. Validation was performed using the 2-tuple linguistic representation model, with eight criteria evaluated by six experts. The architecture comprises four logical layers: Data Acquisition, Communication, Services, and Interaction. These layers work synergistically, encompassing tasks from parameter measurement to user notifications via web or mobile platforms. The results indicate a high level of acceptance, suggesting that the proposed architecture is highly suitable for improving water quality management in aquaculture systems.