Due to advances in additive manufacturing, structures in the mesoscale inspired by nature can be fabricated with reasonable quality and may be used as load-bearing components in industrial engineering sectors. Recently, 3D strut-based lattices inspired by metallic atom set-up have gained more importance since they provide outstanding specific mechanical properties and high energy absorption. Therefore, they may replace conventional structures, e.g., honeycombs as cores in sandwich panels. Besides the excellent mechanical performance of strut-based lattices, the lattice core may enable a multifunctional use of the core by allowing a fluid transfer through the sandwich core. However, designing these kinds of structures as cores in sandwich panels requires experienced knowledge of the lattice mechanical behavior under several load cases. Mainly, the finite element method is used to model the lattice and determine the stresses in the lattice struts, or experiments are performed to observe the failure modes. However, these two methods are time-consuming and may cause inflated costs. Therefore, an analytical model to determine stresses in the struts of 3D lattice cores is introduced in this study. The derived model is based on the homogenization of the lattice to an anisotropic continuum material. Using a higher-order sandwich theory, displacements are obtained in the homogenized lattice core. To assess the stresses of the lattice struts, the homogenized core is replaced anew by the lattice structure using the already obtained displacements (dehomogenization). This kind of modeling reduces the effort to obtain the lattice strut stresses in comparison to FE analysis and thus, simplifies the design process of lattice structures. Furthermore, the method to homogenize and dehomogenize the lattice is applicable to other structures and load cases.

Topology optimization is a powerful tool to automatically generate optimal geometries for additive manufacturing. In topology optimization with SIMP, intermediate densities are typically penalized by locally reducing the stiffness to weight ratio. This avoids grey materials and results in black and white designs that are easy to manufacture. However, topologies with grey-scale-densities may have various advantages, for example a higher stiffness for the same mass. In this paper, structural optimizations of plate-like components are considered as the “variable thickness sheet” problem, where the design variables, i.e., the local densities, represent the thickness of the sheet. To ensure manufacturability by additive manufacturing, a minimum thickness is to be maintained. This is accomplished by a simple interpolation scheme that only penalizes densities between zero and a critical density. For the remaining densities, no penalization is applied. The effectiveness of the approach is demonstrated with two example problems: First, a cantilever beam is optimized, confirming that less penalization produces stiffer structures. Second, an L-shaped bell crank that is to be additively manufactured is optimized. The result is 12% stiffer (and only 4% less stiff) than the design based on conventional (and no) penalization. In regions of load concentrations of plate-like structures, where stresses vary significantly, the results hint at a general potential for performance improvement, when switching from conventional designs (e.g. sheet metals, or standard profiles) to advanced manufacturing methods, such as additive manufacturing.

The Design for Additive Manufacturing of final products needs to target many design objectives, e.g. function, high quality, low lead time, costs, ecological footprint and possibly more. In balancing the latter, results of design changes are frequently counterintuitive, and this happens especially when it comes to sustainability-related qualities. The latter are commonly modeled by a Life-Cycle Assessment (LCA) of all relevant processes of fabrication and use-phase of a product. However, it is likely that related data is not yet available before most design decisions have been fixed. Even though Additive Manufacturing can enable more design freedom than conventional technologies, the production process-specific parameters obviously need strong consideration. The earlier this happens in the design process, improvements might have the biggest impact. Topology optimization can be a very efficient method for conceptual design and design automation in the early design phase. However, the respective models often need to be simplified, e.g. regarding nonlinear material properties or intricate manufacturing constraints. For this reason, it is typically not possible in topology optimization to deal with all previously mentioned criteria.

A Sustainability-oriented Topology Optimization method is proposed within a generative engineering framework. The latter approach is based on interconnection of design processes in an automated way. Multimodal analysis of intermediate topology results should enable the computation of intricate measures. For example, regarding Laser Powder Bed Fusion-based Additive Manufacturing expedient build directions and an estimation of support structures can be calculated. A slim LCA model is included that calculates the sustainability measures on basis of the intermediate topology and reanalysis results. For that purpose, a simplified product system with representative manufacturing and use phase processes is modeled with averaged data. The presented approach enables a more holistic Design for Additive Manufacturing that can deal with a multitude of multidisciplinary criteria in a coherent way.