Optimizing Production Forecasting in Dynamic Complex Industries with Deep Reinforcement Learning Approaches
- 6 December 2025
- journal article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 36 (05) , 697-717
Abstract
Modern factories aim to boost production to increase profits and remain competitive amid intense market pressure. This goal is pursued through production planning, which not only improves process efficiency but also ensures product quality and long-term market success. Among the key tools that support this objective is production forecasting, which plays a central role in facilitating manufacturing processes. In this study, we focus on enhancing the efficiency of industrial systems by developing smart predictive models based on Deep Reinforcement Learning (DRL). Specifically, time-series data from the rotary cement kiln at the SCIMAT factory (Batna, Algeria) were used to predict clinker production. The proposed model, based on the DDQN network, outperformed both the traditional Q-learning network and the DQN. The DDQN achieved a Mean Squared Error (MSE) of 0.0010, a Root Mean Squared Error (RMSE) of 0.0316, and a coefficient of determination ([ ]) of 0.9484. In contrast, the DQN achieved an MSE of 0.0014, an RMSE of 0.0374 and an [ ] of 0.9374. The traditional Q-learning network achieved an MSE of 0.0052, an RMSE of 0.0725 and an [ ] of 0.8865. The LSTM algorithm achieved an MSE of 0.0069, an RMSE of 0.0830 and an [ ] of 0.6570. The developed model is designed to be integrated into the plant’s control system via Programmable Logic Controllers (PLCs), enabling automated operation without direct human intervention. The DQN model demonstrated strong predictive accuracy, leading to improved resource management and operational efficiency. Its application can also be extended to other industrial domains.Keywords
This publication has 26 references indexed in Scilit:
- A constrained multi-objective deep reinforcement learning approach for temperature field optimization of zinc oxide rotary volatile kilnAdvanced Engineering Informatics, 2023
- Reinforcement learning for industrial process control: A case study in flatness control in steel industryComputers in Industry, 2022
- Development of an efficient cement production monitoring system based on the improved random forest algorithmThe International Journal of Advanced Manufacturing Technology, 2022
- Predictive maintenance system for production lines in manufacturing: A machine learning approach using IoT data in real-timeExpert Systems with Applications, 2021
- Parallel spatio-temporal attention-based TCN for multivariate time series predictionNeural Computing and Applications, 2021
- Predictive maintenance in the Industry 4.0: A systematic literature reviewComputers & Industrial Engineering, 2020
- Improvement of a cement rotary kiln performance using artificial neural networkJournal of Ambient Intelligence and Humanized Computing, 2020
- Data driven modeling and nonlinear model predictive control design for a rotary cement kilnIFAC-PapersOnLine, 2019
- From Reinforcement Learning to Deep Reinforcement Learning: An Overview2018
- Model predictive control of a rotary cement kilnControl Engineering Practice, 2011