<![CDATA[This study focuses on optimizing the CNC plasma cutting process for Stainless Steel 410 alloy, a type of martensitic stainless steel known for its high strength, hardness, and good corrosion resistance, using Response Surface Methodology (RSM) combined with advanced optimization methods. Key input variables, including cutting current, speed, and torch height, were systematically analyzed to enhance cutting efficiency by minimizing machining time and surface roughness while maximizing material removal rate. The experimental approach utilized a Central Composite Design (CCD) with 18 trials and Analysis of Variance (ANOVA) to validate linear models as optimal predictors for response variables. Results indicate that cutting speed significantly influences machining time and material removal rate, while cutting current and torch height also influence surface roughness. To improve prediction accuracy and explore the parameter space beyond experimental trials, Artificial Neural Networks (ANN) were implemented, demonstrating superior predictive capabilities compared to RSM alone. Moreover, Multi-Objective Genetic Algorithm (MOGA) and Multi-Objective Ant Lion Optimizer (MOALO) were employed to refine optimization outcomes, addressing trade-offs between conflicting objectives. The optimal configuration, identified as a cutting current of 85.31 A, a speed of 1500 mm/min, and a torch height of 5 mm, achieved a machining time of 0.665 minutes, a surface roughness of 1.77 μm, and a material removal rate of 77.08 g/min. These findings underscore the effectiveness of integrating statistical methods with machine learning and advanced optimization algorithms for precision manufacturing. The study offers a comprehensive framework for improving process efficiency and quality in CNC plasma cutting, catering to the growing demand for high-performance cutting techniques in the steel industry.]]>
