Generative modeling in smart manufacturing: Applications, challenges, and future directions

摘要

Generative modeling has emerged as a transformative tool in smart manufacturing, leveraging advanced machine learning architectures to optimize various manufacturing processes. In the data-rich environment of smart manufacturing, generative models enable the generation and synthesis of diverse process data that drive applications in design automation, quality control, and predictive maintenance. While the potential benefits are substantial, adopting generative models presents several challenges, including model interpretability issues, data privacy concerns, and integration difficulties with existing industrial systems. Addressing these obstacles is essential for the broader implementation of generative modeling across industrial settings. This review systematically explores the applications of generative modeling in smart manufacturing, focusing on its impact in areas such as additive manufacturing, defect prediction, and adaptive supply chain management. Furthermore, by exploring core components of smart manufacturing, this study situates generative modeling within Industry 4.0 contexts, providing a structured overview of recent advancements and technological foundations. This paper highlights future research directions, proposing integrations with digital twins, cyber-physical systems, collaborative robotics, and circular economy to strengthen adaptability, resilience, and real-time decision-making capabilities. By analyzing current applications, challenges, and future pathways, this review aims to equip researchers and practitioners with critical insights, guiding the effective deployment of generative modeling to drive a resilient, data-driven future in smart manufacturing.