The water distribution network carries drinking water across rivers, lakes and water towers to industries, residential users with the help of a complex underground pipelined network system. Checking water defilement or impurities in the water is a significant worry in the field of water circulation systems. The proposed work relies upon the improvement of machine learning and fuzzy system dynamic information with the Internet of Things gadgets to screen the water quality checking structure which uses to screen the boundaries of water, for example, pH, Turbidity and flow rate. Random forest uses a bagging model which is a machine learning supervised learning algorithm which builds and merges multiple decision trees to produce an accurate result in the screening of water quality. To make the framework progressively productive we utilize fuzzy since fuzzy system can give moderate chance outcomes among yes and no, it remains as a decent thinking and dynamic technique like human thinking. In accordance of the water tainting level in the conveyance pipeline, the drinking water quality is named satisfactory/worse/attractive. The framework will caution the client by methods for an email framework to when there is a variation in the water quality boundary from the given predefined edge esteem. This proposed model safeguards standard quality water to the residential people using low prices embedded devices like Launch Pad microprocessor and sensors unit.