Hybrid NSGA-II and deep reinforcement learning framework for adaptive and sustainable turning in industry 4.0
International Journal of Advanced Manufacturing Technology, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s00170-026-18915-8
- Dergi Adı: International Journal of Advanced Manufacturing Technology
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, IBZ Online, Compendex, INSPEC, DIALNET, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO), Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
- Anahtar Kelimeler: Deep reinforcement learning, Industry 4.0, Multi-objective optimization, NSGA-II, Turning process
- Yıldız Teknik Üniversitesi Adresli: Evet
Özet
The development of intelligent machining systems is a key requirement for the transition toward Industry 4.0. This study proposes a hybrid artificial intelligence framework for adaptive turning optimization by integrating statistical analysis, predictive modeling, multi-objective evolutionary optimization (NSGA-II), and deep reinforcement learning. A comparative machinability investigation was conducted on 16MnCr5 alloy steel and AISI 316 L stainless steel under dry and lubricated cutting conditions using a Taguchi-based design of experiments. Surface roughness, cutting temperature, tangential cutting force, and tool wear were analyzed to evaluate machining performance. Response Surface Methodology and Artificial Neural Networks were employed as predictive models, while NSGA-II generated Pareto-optimal solutions representing trade-offs between machining quality and productivity. The resulting optimal solutions were used to guide a deep reinforcement learning agent to ensure safe and efficient adaptive control during machining. Experimental and simulation results demonstrate that the proposed NSGA-II–guided reinforcement learning strategy improves surface quality stability, reduces thermal loads, and slows tool wear progression compared with conventional optimization approaches. The proposed framework provides a scalable cyber-physical solution for intelligent and sustainable machining in Industry 4.0 environments.