A Systematic Investigation of Metaheuristic Algorithms in Smart Manufacturing through Topic Modeling and Fuzzy C-Means Clustering

The development of smart manufacturing has required the incorporation of advanced computational methods to address growing system complexity, variability, and the demand for real time decision-making. This study conducts a systematic investigation into the application of metaheuristic algorithms in smart manufacturing research. A dataset of $\mathbf{3 5 8}$ academic publications is retrieved from the Scopus database, focusing on a wide array of metaheuristic approaches. Nonnegative Matrix Factorization (NMF) is applied to uncover underlying topics, followed by Fuzzy C-Means (FCM) clustering to identify thematic clusters. Six major research topics are identified, including scheduling optimization, predictive machining, cloud manufacturing, robotic logistics, digital manufacturing systems, and sustainable energy-efficient operations. Subsequent clustering analysis reveals seven central clusters, each represented by a core publication, and mapped to relevant topics. The findings reveal a strong emphasis on scheduling and path planning within the literature, with emerging interest in cloud-based service composition and sustainability. This study contributes a novel text mining-based framework for mapping thematic developments in smart manufacturing and highlights key directions for future research.