The use of Additive Manufacturing (AM) can bring opportunities for industry, but several challenges need to be addressed, specifically the digital infrastructure comprising the AM value chain. A combination of a systematic literature review and an industrial use case study concludes that there is low consideration of the digital infrastructure in Design for Additive Manufacturing (DfAM) methods and tools which has a negative impact on the industrialisation of AM. It is therefore recommended that further studies are to be made on how to manage the digital infrastructure in DfAM processes.

Laser Powder Bed Fusion is the most widespread additive manufacturing process for metals. In literature, there are several analytical models for estimating the manufacturing cost. However, few papers present sensitivity analyses for evaluating the most relevant product and process parameters on the production cost. This paper presents a cost model elaborated from previous studies used in a sensitivity analysis. The most relevant process parameters observed in the sensitivity analysis are the 3D printer load factor, layer thickness, raw material price and laser speed.

Laser Powder Bed Fusion (LPBF) brings the possibility to manufacture innovative near-net-shape part designs. But unfortunately, some designed surfaces suffer from rough surface finish due to characteristics of the LPBF process. This paper explores trends in managing surface roughness and through a space industry case study, a proposed process that uses Additive Manufacturing Design Artefacts (AMDAs) is used to investigate the relationship between design, surface roughness and fatigue. The process enables the identification of design uncertainties, however, iterations of AMDA's can be required.

The materials most commonly used in 3D-printers are in a filament form. This is a barrier for users who want to have new types of filaments with different material compositions. A 3D-printer which can extrude and print directly from the raw material was assembled. Compounding with the common additive types; fibres, and metal powders was performed. The size and volumetric output of an extruder was scaling down. Verification was done by mechanical testing, and electron microscopy. The positive result is opening the path to a more accessible composites for both researchers and home producers.

The potentials of additive manufacturing for objectives such as lightweight construction are not yet fully exploited. In this paper, the possibilities of integrative function and system modelling for this challenge will be discussed. In a design study, a triathlon trailer is designed considering the constraints of AM. A suitable system boundary is to be detected using the one-part device method. The findings of the study will help to understand in which form methods such as functional modelling can be applied or adapted for the application of additive manufacturing.

AM-meso structures offer a high potential for adapted properties combined with lightweight design. To utilize the potential a purposeful design of the meso structures is required. Therefore, this contribution presents an approach for modelling their properties depending on design parameters by scaling relationships. The relationships are investigated based on grey box and axiomatic models of elementary cells. Exemplary the pressure stiffness is determined using FEM in comparison to an analytical approximation. The comparison reveals effects and influences occurring within the elementary cell.

In this paper, a proposal for a new method to design load-adaptive microlattice structures for PBF-LB/M manufacturing is presented. For this purpose, a method was developed to stiffen microlattice structures in particular by using self-similar sub-cells to ensure their manufacturability. The quality of the stiffness increase was investigated and verified by finite element simulations. Subsequently, the simulation results were critically discussed with respect to their potential for future design processes for architected materials.

Additive manufacturing (AM) technologies enable the design of new products due to their potentials. The potential of function integration can be extended through a combination of AM with a component integrating technology forming a hybrid integrating additive manufacturing technology (hiAM). With a created development method optimization areas within a product are identified on a functional level using characteristics, structural configurations and integrated functional areas. These are derived analysing examples in literature. The method is applied to a mechanical arm and hand prosthesis.

With the dissemination of additive manufacturing (AM), numerous methods have emerged to support the design process. One possibility is to improve functional solutions through AM-conformal design. Literature-based criteria for the assessment of AM-conformity already exist. Within our study, we address the gap in criteria between a theoretical perspective and a practitioner's perspective. To this end, we first explain the application of the criteria through a use case and conduct an evaluation in an industrial environment adding practitioner's criteria to enable the assessment of AM-conformity.

The increased development in automated driving systems (ADS) has opened up significant opportunities to revolutionize mobility and to set the path for technologies, such as electrification. The proposed methodology is a simulation model backed by a multi-objective optimization algorithm. This research investigates the adoption of future technologies in earthmoving application and explores its implications on the design of future machine concepts in terms of equipment size. The shift from “elephant to ants” in the machine selection, resulted in improved feasibility.

Digital engineering is increasingly established in the industrial routine. Especially the application of machine learning on geometry data is a growing research issue. Driven by this, the paper presents a new method for the classification of mechanical components, which utilizes the projection of points onto a spherical detector surfaces to transfer the geometries into matrices. These matrices are then classified using deep learning networks. Different types of projection are examined, as are several deep learning models. Finally, a benchmark dataset is used to demonstrate the competitiveness.

Finding redundant requirements or semantically similar ones in previous projects is a very time-consuming task in engineering design, especially with multilingual data. Due to modern NLP it is possible to automate such tasks. In this paper we compared different multilingual embeddings models to see which of them is the most suitable to find similar requirements in English and German. The comparison was done for both in-domain data (requirements pairs) and out-of-domain data (general sentence pairs). The most suitable model were sentence embeddings learnt with knowledge distillation.

Design of software in the automotive domain often involves simulation to allow early software parametrization. Modeling complex systems or components impacted by the software in an analytical way can be time-consuming, require domain knowledge and executing the analytical models can result in high computational effort. In specific applications, these challenges can be overcome by applying machine learning based simulation. This contribution presents results of a case study in which powertrain components are modeled data-driven with artificial neural networks to support design space exploration

For successful human-artificial intelligence (AI) collaboration in design, human designers must properly use AI input. Some factors affecting that use are designers’ self-confidence and competence and those variables' impact on reliance on AI. This work studies how designers’ self-confidence before and during teamwork and overall competence are associated with their performance as teammates, measured by AI reliance and overall team score. Results show that designers’ self-confidence and competence have very different impacts on their collaborative performance depending on the accuracy of AI.

The goal in system-level design is to generate a diverse set of high-performing design configurations that allow trade-offs across different objectives and avoid early concretization. We use deep generative models to learn a manifold of the valid design space, followed by Monte Carlo sampling to explore and optimize design over the learned manifold, producing a diverse set of optimal designs. We demonstrate the efficacy of our proposed approach on the design of an SAE race vehicle and propeller.

The present analytical design of shrink fits typically results in an uneven stress condition that can lead to failure in a variety of manners. With increasing loads and the use of brittle materials, the optimization of the stresses in the shrink fit hence becomes increasingly necessary. Currently existing approaches do not solve the problem satisfactorily or increase the manufacturing and design effort. This paper therefore considers the implementation of an AI-based stress optimization using reinforcement learning, which performs stress optimization by geometrically contouring the interstice.

The ongoing digitization of engineering processes and the increasing prevalence of smart products create possibilities for new business models and services. Digital twins enable the collection of all required data about a smart product in order to make these possibilities a reality. This paper describes a flexible approach towards a digital product twin that is tightly integrated with existing product models while being lightweight and easy to integrate with existing IT solutions.

A digital twin (DT) relies on a detailed, virtual representation of a physical product. Since uncertainties and deviations can lead to significant changes in the functionality and quality of products, they should be considered in the DT. However, valuable product properties are often hidden and thus difficult to integrate into a DT. In this work, a Bayesian inverse approach based on surrogate models is applied to infer hidden composite laminate ply angles from strain measurements. The approach is able to find the true values even for ill-posed problems and shows good results up to 6 plies.

This work studies the perception of the impacts of AI and human process managers during a complex design task. Although performance and perceptions by teams that are AI- versus human-managed are similar, we show that how team members discern the identity of their process manager (human/AI), impacts their perceptions. They discern the interventions as significantly more helpful and manager sensitive to the needs of the team, if they believe to be managed by a human. Further results provide deeper insights into automating real-time process management and the efficacy of AI to fill that role.

Engineers widely use topology optimization during the initial process of product development to obtain a first possible geometry design. The state-of-the-art method is iterative calculation, which requires both time and computational power. This paper proposes an AI-assisted design method for topology optimization, which does not require any optimized data. The presented AI-assisted design procedure generates geometries that are similar to those of conventional topology optimizers, but require only a fraction of the computational effort.