Forecasting electricity consumption of Malaysia’s residential sector: Evidence from an exponential smoothing model

Abstract

Background: In Malaysia, residential electricity consumption has shown a steady upward trend year by year. Due to this increase in energy consumption, it is important to forecast the value of electricity consumption until the year 2032 to accommodate the electricity demand. Methods: Three exponential smoothing models were compared to identify the most appropriate model in forecasting electricity consumption. The three exponential smoothing models are Simple, Holt, and Brown exponential smoothing. To identify the most appropriate model, a mean absolute percentage (MAPE) was chosen. Results: The results show that Holt’s exponential smoothing has the best performance with the lowest MAPE score of 2.299. Conclusions: Consequently, it was found that electricity consumption would substantially increase from 2647 ktoe (kilotonne of oil equivalent) to 3873 ktoe within the period of 2019 to 2032.