Synthetic data generation in smart manufacturing applications: A systematic review and outlook

Abstract

Data-driven methods, particularly those based on machine learning and artificial intelligence, are transforming decision-making in smart manufacturing. However, these methods typically require large volumes of high-quality labeled data for training, which are often difficult to obtain due to limited availability, class imbalance, privacy concerns, and the high costs and time demands of data collection. Synthetic data generation (SDG) has emerged as a promising approach to address these challenges. Over the past two decades, SDG techniques have evolved through diverse paths, including classical data augmentation, rule-based methods, simulation-based methods, and deep generative methods. Despite these advances, the adoption of SDG in manufacturing is still in its nascent stage, with applications mostly confined to proof-of-concept demonstrations and narrowly scoped use cases. Furthermore, a comprehensive, systematic literature review of SDG in manufacturing applications is lacking. To address this gap, this paper presents a systematic literature review of SDG for smart manufacturing applications, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Drawing on 201 identified publications spanning 2018 to early 2025, the review synthesizes the current state of the field and identifies key opportunities for future research. Viewed through the lens of the product lifecycle, the findings demonstrate how SDG can serve distinct functions at each stage of manufacturing, thereby revealing opportunities for advancing its integration into various smart manufacturing practices.

The authors' abstract, as published at the source. Journal of Manufacturing Systems, 2026 · DOI ↗