<![CDATA[Efficient production planning is crucial in the manufacturing industry, including in the paper sector, where fluctuating demand and limited production capacity pose significant challenges. This study introduces an intelligent optimization system that integrates demand forecasting using Long Short-Term Memory (LSTM) with production scheduling optimization through Linear Programming (LP) in Pyomo. The LSTM model processes historical order data to predict demand for the next 30 days, which is then used as input for the LP model to generate an optimal production schedule while considering machine capacity and operational time constraints. The experimental results indicate that the LSTM model achieves a prediction error (loss) of approximately 0.032, demonstrating high accuracy in capturing demand patterns. Meanwhile, the LP model implemented in Pyomo efficiently allocates production time, ensuring that machine utilization is optimized without exceeding the available working hours. By integrating these approaches, companies can minimize the risks of overproduction and stockouts while maximizing resource efficiency. Furthermore, this method enhances decision-making processes by providing data-driven insights into production scheduling and inventory management. The proposed framework offers a scalable solution for improving operational performance in the paper industry, enabling companies to respond more effectively to market fluctuations and optimize their supply chain strategies.]]>
Copyrights © 2025
