Multi-objective optimization for dimensional accuracy and ultimate compression strength of polystyrene (PS200) parts using laser powder bed fusion process

Purpose The Laser Powder-Bed Fusion (L-PBF) process has shown growing potential in fabricating high-quality patterns for casting. This study aims to optimize L-PBF process parameters for Polystyrene (PS200) to enhance dimensional accuracy and Ultimate Compression Strength (UCS), ensuring pattern reliability during handling and molding. Design/methodology/approach A three-factor, three-level Box-Behnken Design under Response Surface Methodology was used to evaluate the effects and interactions of Laser Power, Laser Speed and Hatch Distance on UCS and dimensional deviation. Orientation-based mechanical and geometric evaluations were performed, followed by multi-objective optimization using the Non-Dominated Sorting Genetic Algorithm (NSGA-II). Findings Dimensional accuracy varied with orientation, with the Y-direction showing the lowest deviation (0.04–0.17 mm), while UCS was highest along the X-direction (1.16–3.85 MPa), indicating directional dependency. Optimization using NSGA-II provided a set of trade-off solutions, and the selected parameter set was used to fabricate a spherical PS200 pattern. The final aluminum cast part exhibited less than 1% dimensional deviation from the original pattern, validating the effectiveness of the optimized process. Research limitations/implications The findings are based on a specific machine–material combination. Broader generalization requires further research across different machines, geometries and material variants. Originality/value This study validates the use of PS200 in L-PBF for vacuum casting, achieving optimized, reliable patterns for high-precision small-batch production.