<p>This study investigates the mechanical and microstructural responses of spheroidal graphite cast iron under varying process parameters using a statistical framework. An L18 orthogonal array based on the Taguchi method was applied to evaluate the effects of nine selected factors, including carbon equivalent, type and amount of spheroidizer, type and amount of inoculant, inoculation method, pouring temperature, wall thickness, and scrap ratio. Mechanical properties such as tensile strength, elongation, yield strength, and hardness were assessed using signal-to-noise ratios derived from repeated measurements across three section thicknesses. Principal component analysis (PCA) captured over 78% of the variance, with strength-related properties aligned along the first axis and ductility-related responses along the second, confirming the inherent trade-off structure. K-means clustering of PCA scores classified the 18 runs into three distinct regimes corresponding to strength-oriented, ductility-oriented, and balanced conditions, supported by differences in ferrite and pearlite fractions. The integrated application of Taguchi design, robustness evaluation, and multivariate analysis provides a quantitative basis for process optimization in ductile iron casting and offers a transferable framework for data-driven materials design.</p>