PT Olifant, as a company operating in the manufacturing sector, has its address in Pasir Jaya village, Jatiuwung District, Tangerang City, carries out energy management in the form of energy usage planning, namely the process of predicting energy usage using average values from historical data. However, it turns out that it does not have high accuracy, especially if there are data anomalies within that period. To get better accuracy, it is necessary to classify the reference data (training data) so that it can be sorted out which data needs to be used and which data is not necessary. This classification process can be done with Support Vector Machine (SVM), which is one application of Machine Learning. By using SVM, it is proven to have good capabilities on limited historical data. The steps that will be used in this research are collecting electricity consumption data through literature studies of documents or energy consumption records and field observations. The data obtained will be processed using the Support Vector Machine (SVM) method which consists of 2 phases; First, the Training Phase to recognize energy consumption patterns from historical data. The input data will be classified and will be represented in a formula. Second, the Testing Phase is for applying the formula from the training phase to actual data on electrical energy consumption. Predicted data will be tested with actual data to see any deviations that occur. This is a measure of success that is monitored, namely the level of accuracy in predicting electrical energy consumption patterns resulting from the introduction of electrical energy consumption patterns in the training phase.