<![CDATA[Electricity is a vital necessity for society and serves as a key driver across various sectors, including households, businesses, and industries. With the increasing demand for electricity each year, PT. PLN (Persero) ULP Lhokseumawe is required to plan its distribution and power capacity accurately. Inaccurate forecasting may cause imbalances between supply and demand. This study compares two forecasting methods, namely Single Exponential Smoothing (SES) and Double Exponential Smoothing (DES), to predict electricity consumption in the Lhokseumawe region. The dataset consists of monthly electricity consumption per sub-district from 2022 to 2024, with forecasting projections up to 2027. The research stages include data collection, preprocessing, application of SES and DES methods, accuracy evaluation using Mean Absolute Percentage Error (MAPE), and the design of a web-based system using Python and Flask. The results indicate that the SES method achieved higher accuracy with a MAPE value of 5.85%, while the DES method obtained a MAPE value of 7.87%. These findings suggest that SES is more suitable for data with random fluctuations, whereas DES is better applied to data with trend patterns. By comparing the MAPE values, this study provides insights into which method is more optimal for electricity consumption forecasting in Lhokseumawe. The outcomes are expected to contribute practically to PT. PLN (Persero) ULP Lhokseumawe in formulating more effective and efficient electricity distribution strategies.]]>
Copyrights © 2025
