Forecasting Electricity Consumption Using Exponential Smoothing Methods

Despite being a need for daily living, electricity is a scarce resource. It is crucial to create an appropriate model to calculate the forecasting of power demand that is required by decision-makers. The domestic electric consumption has contributed 20.7% of Malaysia's overall energy consumption in recent years. Because of shifting lifestyles Malaysia's domestic electricity consumption is predicted to rise. Thus, the goal of the current study is to forecast domestic electrical consumption using single-exponential and double-exponential methods. Using prior knowledge, these methods determine the smoothing parameters' values. Applying exponential smoothing techniques and seeing the results demonstrates its efficacy in predicting electricity demand. The excellent models are double exponential smoothing method were selected by identifying the forecast errors with the lowest value. Using mean absolute percentage error (MAPE) and mean absolute deviation (MAD) as forecast accuracy. As an end result, double exponential model offers significant results that indicate it is superior to exponential smoothing methods as a forecasting model.