EUVIMEDEUVIMED entdecken
Crossref

Synergistic application of response surface methodology and machine learning for strength optimization in 3D-printed PLA composites

Shashwath Patil

ARPN Journal of Engineering and Applied Sciences · 20.04.2026 · 249

DOI: 10.59018/022635 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ARPN Journal of Engineering and Applied Sciences, online erschienen am 20.04.2026. Verfasst von Shashwath Patil. Die Arbeit nennt 26 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In the current study, polylactic acid (PLA) filaments reinforced by Squid pen powder (SPP) were 3D printed into filaments and prepared through fused deposition modeling (FDM). The key aim was to maximize tensile strength, and this was done by investigating the concomitant influence of the layer thickness, nozzle temperature, and filler ratio. Response Surface Methodology (RSM) applied a central composite design (CCD) to design regression models and to determine statistically significant factors. The results showed that the most significant impact was on the layer thickness, then the filler ratio, but the nozzle temperature had a relatively weak effect. The optimum material parameters calculated by the RSM were identified to be a layer thickness of 0.2 mm, a nozzle temperature of 220 °C, and a filler ratio of 6 wt.%, which gave the highest tensile strength. To increase predictive accuracy, machine learning (ML) models, such as Support Vector Regression (SVR), Random Forest (RF), and Multi-layer perceptron (MLP) models, were developed. RF performed better, as was supported by a feature importance analysis, which again showed the prevalence of layer thickness. Furthermore, Bayesian optimization combined with RF found similar optimal scenarios and provided a slightly better predictive quality. The integrated RSM-ML system, therefore, provides a powerful tool in the optimization of the tensile strength of 3D-printed polymer composites.

Zitieren

APA

Patil, S. (2026). Synergistic application of response surface methodology and machine learning for strength optimization in 3D-printed PLA composites. ARPN Journal of Engineering and Applied Sciences, 249. https://doi.org/10.59018/022635

Vancouver

Patil S. Synergistic application of response surface methodology and machine learning for strength optimization in 3D-printed PLA composites. ARPN Journal of Engineering and Applied Sciences. 2026:249. doi:10.59018/022635

BibTeX
@article{Patil2026,
  author = {Patil, Shashwath},
  title = {Synergistic application of response surface methodology and machine learning for strength optimization in 3D-printed PLA composites},
  journal = {ARPN Journal of Engineering and Applied Sciences},
  year = {2026},
  pages = {249},
  doi = {10.59018/022635}
}
Kostenlos starten →