Inland fisheries provide food and livelihood to 1.9 million Filipinos, contributing 51.8% of national fisheries production. This sector has positioned the Philippines as the sixth-largest aquaculture producer globally. However, recurring fish kills in Philippine lakes during 2023-2024 have highlighted critical water quality management challenges. This study enhances the sustainability of inland fisheries by developing a real-time water quality monitoring system with IoT technology that sends data to the backend for data analytics integration. The system achieved comparable accuracy to commercial multiparameter devices. A developmental research design was employed to develop a system that could be used in effective water quality monitoring. A quantitative research method was used in data collection. The system monitors the water quality parameters such as pH, dissolved oxygen, and temperature. Machine learning models were developed for predictive analytics to forecast water quality trends. Prescriptive analytics have been added to recommend actions when the water quality status is not normal. Water samples from different lakes were used in testing the system and ISO/IEC standards were used during the evaluation. The system achieved 99.39% accuracy for temperature, 94.97% for pH, and 80.67% for dissolved oxygen measurements.

Keywords: IoT-based water quality monitoring; predictive analytics; inland fisheries sustainability; machine learning models; Philippine aquaculture