The forecast load requirements for electrical energy are an important fist step in planning and developing electricity supply at any time sufficiently, well dan continously. Therefore we need a load forecasting method that is accurate and easy to implement based on available data on the Autoregrsive Integrated Moving Average (ARIMA) method. So the advantages in this ARIMA Method are good for short-term forecasting, flexible and can represent a wide range of time series characters that occur in the short term, there are formal procedures in testing the suitability of the model and forecast interval and predictions have followed the model. Period of data in school by clustering the monthly data, from the results of cluster clustering forecasting the burden of each monthly period in the future can be done