Experimental investigation and statistical modeling of surface roughness in conventional drilling of Al-2024/SiC composites

Purpose This study aims to systematically investigate the effects of drilling parameters and SiC reinforcement content on the surface roughness of Al-2024 aluminum matrix composites fabricated by stir casting. The study further seeks to determine the optimum drilling conditions using the Taguchi optimization method, quantify the statistical significance of the machining parameters through analysis of variance (ANOVA) and develop a reliable regression model for predicting surface roughness. The ultimate objective is to provide practical machining guidelines for improving the surface integrity of particle-reinforced aluminum composites used in high-performance engineering applications. Design/methodology/approach Unreinforced Al-2024 alloy and composites reinforced with 10 and 15 vol.% SiC particles were fabricated using the stir casting technique, followed by T6 heat treatment. The microstructures of the produced materials were characterized by optical microscopy and scanning electron microscopy (SEM), while Brinell hardness measurements were performed to evaluate the mechanical response. Drilling experiments were conducted on a CNC milling machine using TiN-coated HSS twist drills according to a Taguchi L27 orthogonal array. Surface roughness (Ra) was measured using a MarSurf VD140 contact profilometer. Statistical analyses, including Taguchi optimization, ANOVA, response surface analysis and multiple linear regression, were employed to evaluate the effects of material type, cutting velocity and feed rate on surface roughness. Findings The results demonstrated that feed rate was the dominant machining parameter affecting surface roughness, contributing 82.85%, followed by cutting velocity (12.93%) and material type (1.93%). The optimum drilling condition for achieving the minimum surface roughness was obtained at a cutting velocity of 25 m/min and a feed rate of 0.05 mm/rev for unreinforced Al-2024. Optical microscopy and SEM analyses revealed that increasing the SiC reinforcement content promoted particle fracture, particle pull-out and surface damage during drilling. In contrast, Brinell hardness increased with increasing SiC content, reaching approximately 95 HB for the 15 vol.% SiC-reinforced composite after T6 heat treatment. Furthermore, the developed regression model exhibited excellent predictive capability with a coefficient of determination of R2 = 95.69%, confirming the reliability of the proposed statistical methodology. Originality/value This study presents a comprehensive experimental and statistical investigation of the conventional drilling behavior of SiC particle-reinforced Al-2024 composites produced by stir casting. Unlike many previous studies that primarily focused on advanced machining techniques, the present work demonstrates that conventional drilling, when combined with systematic statistical optimization, can provide high-quality surface finish with reliable predictive capability. The excellent agreement among the microstructural observations, Taguchi optimization, ANOVA, response surface analysis and regression model provides a robust framework for machining optimization and offers practical guidance for the aerospace and automotive industries.