Shape design of stretching- and bending-dominated lattice columns for Wire-and-Arc Additive Manufacturing

Wire-and-Arc Additive Manufacturing (WAAM) enables the fabrication of lattice meso-elements made of spatial networks of thin branches. In stretching-dominated lattices, members primarily carry loads through axial tension and compression, providing high stiffness-to-weight ratios that are well-suited for lightweight structural components requiring high rigidity. In contrast, bending-dominated lattices deform mainly through flexural actions. Although less efficient in terms of stiffness, they are more deformable and therefore advantageous in other applications. In this contribution, an energy-based computational framework is implemented to automatically design shape-optimized lattice columns for robotic fabrication. The lattice geometry is described through a low-dimensional Hermite-spline parametrization, which is explicitly constrained by WAAM overhang and printability limits, enabling automation-ready shape generation without manual redesign. A minimum compliance optimization procedure is employed to generate stretching-dominated lattice spatial structures, whereas a compliance-constrained maximization of the bending contribution to the overall strain energy is devised to achieve bending-dominated layouts. The numerical studies on meso-columns under central and eccentric loads, together with a WAAM-produced prototype, demonstrate the feasibility of the digital design and fabrication workflow for a new generation of steel structures.