Real-Time Data Forecasting On Missing Energy Data Using Seasonal Autoregressive Integrated Moving Average (SARIMA) Model

Energy data analysis is becoming more important for producing efficient and cost-effective energy by monitoring usage and identifying when energy loses quality. Time series analysis helps predict future trends and supports decision-making. This study uses the SARIMA model to forecast missing energy data from Hospital Rehabilitasi Cheras. The model’s performance was measured using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), showing its ability to handle seasonal data patterns. Real-time energy data (kWh) from the hospital, collected every 30 minutes over 6 months, was analyzed. The best model identified was SARIMA (2,0,0) x (1,1,1)24 for forecasting missing weekday data with seasonal patterns. The model’s parameters were evaluated, achieving an MSE of 2144.94514 and an RMSE of 46.31355. The model’s accuracy was further evaluated by comparing the actual and forecasted values, taking into account the complex correlation of socioeconomic factors affecting energy usage.