A Neuro-Fuzzy Approach for Domestic Water Usage Prediction

The unconstrained rise in water usage as a result of population growth, rapid urbanization and climate change has become an issue of paramount concern for policy makers across the globe. Consequently, fresh water as a renewable but finite resource must be managed efficiently to sustain domestic and productive activities. Efficient water management strategies must be developed to address the challenges of increased demand without undermining long term sustainability. Developing such strategies necessitates a multidisciplinary approach incorporating policy planning and applied technology to efficiently manage water resources for maximizing economic growth and promoting social welfare. Towards this goal, we develop a hybrid intelligent system for domestic water usage prediction based on Adaptive Neuro-Fuzzy Inference System (ANFIS). The proposed system is trained in a supervised manner to model the relationship between environmental factors and domestic water consumption. The system forecasts domestic water usage based on environmental factors particularly atmospheric pressure, temperature, relative humidity and wind speed. Evaluation of the system on a real smart home dataset demonstrates that the system predicts domestic water consumption with higher accuracy.