A Statistical Study on the Effect of Print Speed and Layer Thickness on Yield Strength in Additive Manufacturing

This study quantifies how print speed and layer thickness jointly influence the yield strength of fused deposition modeled thermoplastic polyurethane (TPU) parts. A 3×3 full factorial experiment (print speed: 600, 900, 1200 mm·min⁻¹; layer thickness: 0.10, 0.20, 0.25 mm) with two replicates per treatment (n = 18) was conducted. Dog-bone specimens were tensile-tested per ASTM D638 on a universal testing machine to obtain yield strength. Model assumptions were evaluated; a log₁₀ transformation of the response resolved non-normality and heteroscedasticity. Two-way ANOVA revealed a statistically significant interaction between print speed and layer thickness (p ≈ 0.001), prompting interaction-aware inference. Post-hoc Tukey and Bonferroni comparisons showed that the combination of 600 mm·min⁻¹ with 0.20 mm produced the highest yield strength, whereas 1200 mm·min⁻¹ with 0.25 mm produced the lowest which is an improvement on the order of fivefold between the best and worst settings. No outliers were detected; a slight serial trend in residuals was noted. Results highlight that optimal mechanical performance in FDM depends on factor interactions rather than main effects alone, and they offer actionable guidance slower deposition with mid-thickness layers for parameter selection in TPU printing, alongside a reproducible statistical workflow applicable to other AM materials and responses.