From infill to interface: Taguchi optimization and cohesive zone modelling of adhesively bonded FDM PLA lap joints

Fused deposition modelling (FDM) is increasingly used to fabricate functional polymer parts. However, the limited build volume of FDM printers makes the assembly of subcomponents through adhesive bonding necessary for the production of larger structures. The mechanical performance of such bonded assemblies depends on a coupled interaction between the adherend internal architecture (infill pattern and density) and the mechanical and fracture properties of the adhesive. This study presents a systematic experimental and numerical investigation of adhesively bonded single-lap joints (SLJs) fabricated from fused deposition modelling (FDM) polylactic acid (PLA) adherends. A Taguchi L9 orthogonal array was employed to examine the combined effects of three process parameters—infill pattern (triangular, honeycomb, grid), infill density (20%, 40%, 60%) and adhesive type (Araldite 2015 epoxy, Loctite 330 acrylic, Sikaflex 252 polyurethane)—on peak load (Fmax) and joint stiffness (k). ANOVA results identified adhesive type as the dominant factor influencing Fmax (68.2%), while infill density governed stiffness (45.2%). The honeycomb/60%/Araldite 2015 configuration achieved the highest peak load (1138 N), representing a 134% improvement over the weakest condition. Effective elastic moduli of the PLA adherends, determined via ASTM D638 tensile testing (593–999 MPa), were incorporated into linear elastic and cohesive zone model (CZM) finite element analyses implemented in ANSYS Workbench. CZM predictions agreed with experimental peak loads within ±8.75% for the Araldite 2015 configurations and ±5.75% for the Loctite 330 configurations, demonstrating reliable parameter transferability to FDM PLA structures. The Volkersen shear-lag model was applied analytically to all configurations; stress concentration factors ranged from 1.18–1.24 for polyurethane to 10.71–12.27 for epoxy. Comparison with the FEA results, which showed peel stresses up to −14.18 MPa at the overlap ends, demonstrated the complementary role of FEA in capturing through-thickness stress components that are outside the scope of the Volkersen model. The integrated Taguchi–CZM–Volkersen framework provides an efficient methodology for optimising adhesively bonded FDM polymer assemblies.