Abstract: Industrialization has significantly contributed to economic growth; however, it has also intensified environmental pollution, particularly through the discharge of untreated or partially treated industrial effluents into natural water bodies. These effluents contain chemical contaminants, suspended solids, organic matter, and toxic compounds that severely degrade water quality and threaten ecological balance and human health. Traditional monitoring techniques depend on manual sampling and laboratory-based chemical analysis, which are periodic, labour intensive and incapable of detecting sudden pollution spikes in real time. This research proposes an Intelligent IoT-Based Real-Time Industrial Effluent Monitoring System integrated with Machine Learning (ML) for automated classification of effluent quality. The system continuously measures critical water quality parameters such as Total Dissolved Solids (TDS), Turbidity, Electrical Conductivity and Temperature using calibrated sensors connected to an ESP32 microcontroller. The sensor data are transmitted through wireless communication to a cloud server where preprocessing and classification are performed using a Random Forest model. The classification thresholds are derived from environmental discharge standards established by the Central Pollution Control Board and the World Health Organization. The proposed system not only enables real-time monitoring but also provides intelligent pollution categorization and automated alerts. Experimental results demonstrate high classification accuracy, reduced response time and improved reliability compared to conventional threshold-based systems.

Keywords: IoT, Industrial Effluent Monitoring, Machine Learning, Random Forest, Water Quality Classification, Environmental Pollution and Smart Monitoring Systems

Downloads: | DOI: 10.17148/IJIREEICE.2026.14355