Literature Review: The Role of IoT and AI in Water Quality Monitoring in Aquaculture

Abstract The main challenge in modern aquaculture is meeting increasing production demands while simultaneously facing various environmental pressures. Water quality parameters such as temperature, pH, dissolved oxygen, ammonia, and salinity play central roles because they directly influence organism growth and health. Instability in these parameters may trigger mass mortality and significant economic losses, thereby encouraging the need for reliable monitoring systems (Al Mamun, 2024; Jais, 2024). To overcome the limitations of traditional practices that are manual and not real-time, the Internet of Things (IoT) offers solutions through sensor infrastructures capable of continuous data acquisition and transmission. Field implementations vary, ranging from buoy modules to stationary solar-powered systems that facilitate remote monitoring and cloud-based data visualization (Shete, 2024; Abdillah, 2025). Beyond mere data collection, information value is enhanced through Artificial Intelligence (AI) analytical capabilities. Various Machine Learning and Deep Learning algorithms (such as Random Forest, SVM, and LSTM) have been applied for time prediction, anomaly detection, and automated control (Mahesh, 2024; Alluhaidan, 2025). This IoT–AI synergy promises more proactive management systems. The IoT–AI combination essentially forms a new paradigm in aquaculture management that is precise, adaptive, and efficient. However, several challenges such as sensor durability, network dependency, and lack of data standardization still hinder implementation. This paper aims to further explore IoT and AI applications in water quality monitoring as well as their future development opportunities.