<![CDATA[In the competitive food and beverage industry sector, the ability to accurately predict demand is crucial to supporting effective production and marketing strategies. Chintari Cake and Cookies, a small and medium-sized enterprise (SME) specializing in homemade cakes and cookies, faces challenges in dealing with unpredictable demand fluctuations. This study aims to forecast daily sales using the Long Short-Term Memory (LSTM) algorithm, a type of Recurrent Neural Network (RNN) known for its effectiveness in processing sequential data and recognizing long-term patterns. LSTM was chosen due to its advantages over conventional statistical methods such as ARIMA, particularly in terms of prediction accuracy. Five years of historical sales data were used as model input, which was then processed through preprocessing stages before training the LSTM model. The prediction results were evaluated using RMSE (Root Mean Square Error) and MAPE (Mean Absolute Percentage Error) metrics. The results showed an RMSE value of 6.752 and a MAPE value of 6.792, indicating a low prediction error rate. These findings demonstrate that the LSTM algorithm can serve as an effective solution for SMEs in improving the accuracy of production planning and inventory management based on historical data patterns.]]>
Copyrights © 2025
