<![CDATA[This study compares the performance of the Fuzzy Time Series (FTS) Cheng and Lee methods in forecasting the distribution, production, and sales of clean water at Perumda Tirta Silaupiasa, Asahan Regency. The research utilized monthly secondary data from January 2023 to April 2025. Data analysis was conducted using R-Studio software, and forecasting accuracy was evaluated through the Mean Absolute Percentage Error (MAPE). The results show that both methods achieved a high level of accuracy, with MAPE values below 3%. However, the Cheng method consistently outperformed the Lee method, recording MAPE values of 2.28% (distribution), 1.59% (production), and 1.63% (sales), while the Lee method produced slightly higher values of 2.45%, 1.83%, and 1.77% for the same variables. These findings demonstrate that the Cheng method is more adaptive in capturing fluctuations in clean water data, especially during periods of sudden changes. This research provides practical recommendations for improving clean water supply management in regions with dynamic operational conditions]]>
Copyrights © 2025
