Water is a natural resource that needed by all living things. Humans, animals, and plants need water to survive. Unlike animals and plants, humans need clean water to survive. Becoming a challenge for all PDAM Indonesia to keep sufficient number of demand for clean water supply, not to mention PDAM Malang City. Prediction done in this research use Exponential Smoothing method. Exponential Smoothing is a method that continuously performs forecasting improvements by taking the average value of smoothing past values from time expanding data in exponential way. In this research, we will compare three Exponential Smoothing methods: Single Exponential Smoothing (SES), Double Exponential Smoothing (DES), and Triple Exponential Smoothing (TES) used to obtain prediction result and evaluate prediction result with Mean Absolute Percentage Error (MAPE). The smallest MAPE was obtained when using Single Exponential Smoothing (SES) method when value ɑ = 0.2 with MAPE value 3.992, Double Exponential Smoothing (DES) method when value ɑ = 0.1 with MAPE value 4.932, and Triple Exponential Smoothing method TES) when the value of ɑ = 0.1, β = 0.1, and γ = 0.6 with MAPE value of 6.733. With the MAPE value below 10, the Exponential Smoothing method to predict the amount of water requirement included into the category is very good.
                        
                        
                        
                        
                            
                                Copyrights © 2018