Normal boundary intersection method based on principal components and Taguchi’s signal-to-noise ratio applied to the multiobjective optimization of 12L14 free machining steel turning process

Abstract

When it comes to multiobjective optimization problems, the challenge is to find a solution that satisfies all the answers simultaneously. When the responses are correlated and present conflicting objectives, it is even more difficult to find an adequate solution, since most optimization techniques do not consider this information. The objective of this work is to apply Taguchi’s signal-to-noise ratio (SNR) and principal component analysis (PCA) in order to standardize the optimization objectives, eliminate the correlation between the multiple responses, and combine them with the normal boundary intersection (NBI) method to perform a proper optimization. A case study of 12L14 free machining steel turning process is used, since it is considered an important machining operation. Three input parameters (cutting speed, feed rate, and depth of cut) and three response variables (mean roughness, total mean roughness, and removal rate) were considered. Response surface methodology was employed to build the objective functions. The NBI-PCA-SNR method was successfully applied, presenting viable solutions.