Fossil fuel is one of the most important energy sources in Indonesia and is widely used both in transportation, industry, and households. Fossil fuel is a fuel formed by natural processes such as the anaerobic decomposition of buried dead organisms and their resulting fossil fuels typically have an age of millions of years. Energy consumption in Indonesia, based on data from EIA (Energy Information Administration) from 1980 to 2018, has been rising sharply as a result of the high population. This condition is certainly a serious concern because if new energy reserves are not found, Indonesia is going to have an energy crisis.
Forecasting primary energy production and consumption markets is critical for efficient energy policy implementation. More precise forecasts of energy consumption and production are vital when consumption growth rates are greater than production growth rates as in the case of Indonesia. To understand the future of energy consumption and production, we utilize machine learning algorithms, Holts Winter forecasting, Seasonal Autoregressive Integrated Moving Average (SARIMA), and Autoregressive Integrated Moving Average (ARIMA) model. We trained and evaluated the model using annual data from 1980 to 2018. The Holts Winters model and SARIMA outperform the ARIMA. The prediction shows that the supply of dry natural gas and coal is secure, whereas the supply of petroleum is in jeopardy. Coal is the safest energy source, according to forecasted predictions, with considerable production, growing 23% from 2018.