Optimization of Cnc Machining Parameters for Improved Surface Quality and Productivity

Abstract

This paper presents a reproducible experimental methodology for simultaneously improving surface quality and productivity in CNC turning. Cutting speed, feed rate, and depth of cut are varied at three levels through a Taguchi L27 orthogonal array. Average surface roughness (Ra) is treated as a smaller-the-better response, whereas material-removal rate (MRR) is treated as larger-the-better. Signal-to-noise ratios, grey relational analysis (GRA), analysis of variance (ANOVA), regression modeling, and a confirmation run are integrated so that conflicting responses can be optimized with a single experimental plan. AISI 1045 steel and a coated-carbide insert are specified as the reference system. The worked results indicate that feed is expected to dominate Ra, while depth of cut and feed dominate MRR; a balanced parameter setting can therefore raise throughput without allowing roughness to exceed an engineering limit. Numerical results are illustrative until replaced by measurements from the proposed experiment. The method is suitable for laboratories and small manufacturing units seeking an economical, statistically defensible parameter window.