Principle Component Analysis and Social Network Analysis for Decision Support of Ultra-Precision Machining

Ultra-precision machining (UPM) technology is actively engaged in manufacturing high technological products nowadays. However, as its complicated machining mechanisms, the induced intricate relationships between various machining factors erect barriers in obtaining optimal machining conditions. The goal of this study is to use social network analysis (SNA) and principal component analysis (PCA) to combine the metrics of individual UPM factors and prioritize UPM factors based on their combined characteristics. In the beginning, the preliminary results of SNA approach act as the input to conduct PCA and generate principal components (PCs). The PCs were then combined into multiple characteristic performance indexes (MPCI), which have the balance characteristics of all main metrics from SNA, allowing to demonstrate the UPM factors with relatively high MPCI to be the dominant variables in optimizations. Few case studies have been provided for validation of the effectiveness of adjustments in UPM factors with high MPCI on the machining outcomes. The optimal machining conditions with multi-objectives could be effectively reached by executing the machining strategies with considering the prioritized UPM factors from the results of the hybrid SNA and PCA approach in this study. Overall, this study contributes to providing a comprehensive reference to academics and industry for prioritizing UPM factors with considering the balanced machining outcomes and developing practical machining strategies.