Vision-Guided Process Optimization for Stringing Reduction in FDM 3D Printing Using RSM

Abstract Fused Deposition Modeling (FDM) often faces problems in the quality of the parts due to formation of defects such as stringing, dimensional inaccuracy and warping. Stringing is a defect in FDM, where tiny strings of plastic material exist on the printed objects. This study employs the Central Composite Design (CCD) approach under Response Surface Methodology (RSM) to conduct experimental trials, optimizing critical process parameters influencing stringing, which includes nozzle temperature, retraction speed, layer thickness and print speed. A vision-based defect identification approach is employed to detect and quantify the presence of stringing in printed parts. A regression model is developed to analyze the effect of process parameters on the formation of stringing. The optimal parameters were determined as nozzle temperature of 198◦C, retraction speed of 36 mm/s, print speed of 145 mm/s and layer thickness of 0.14 mm. These findings are crucial for manufacturing industries where precision and reliability are vital. By addressing the stringing in FDM processes, this work enhances the quality of printed parts while reducing reliance on post-processing, thereby promoting broader industrial adoption of FDM technology.