Hybrid Taguchi-GRA-PCA approach for multi-response optimisation of turning process parameters under HPC condition

Abstract

Process parameter optimisation is important in manufacturing. An effective and efficient manufacturing process offers low temperature, low force and high material removal rate without compromising the product quality. An experimental investigation designed by Taguchi's orthogonal array for turning AISI 1060 steel by uncoated carbide insert under high-pressure coolant (HPC) condition. Since all these aforementioned responses are mutually conflicting, simultaneous multi-response optimisation is required to acquire a trade-off among the responses. Standalone Taguchi method is not sufficient for multi-response optimisation. Grey relational analysis (GRA) can be coupled with Taguchi method to overcome this complexity, where overall process performance index grey relational grade (GRG) is calculated and used for the optimisation of the process by using Taguchi's S/N ratio. Principal component analysis (PCA) can be utilised to determine the individual importance of each response, which facilitate accurate computation of overall GRG. The effect of process parameters on GRG was analysed using analysis of variance (ANOVA). Validation by confirmation test was also performed.