This study benchmarks multiple data-driven methodologies for predicting relative density (RD) of 316 L stainless steel fabricated via Powder Bed Fusion–Laser Beam (PBF-LB), as part of the ESAFORM Benchmark 2025 AMDmodel initiative. Two datasets (DS-01 and DS-02), each with 256 specimens from a 4-factor, 4-level design of experiments, were produced on different PBF-LB systems equipped with equivalent in-situ infrared (IR) melt-pool pyrometry. Failed builds (RD = 60

Purpose The layered construction inherent to additive manufacturing (AM) processes introduces unique challenges related to part geometry. This demands a better integration of knowledge during design to avoid the large number of design iterations. This paper aims to establish a framework for effectively delivering AM-related knowledge throughout the design process, particularly for designers who are not AM experts. Design/methodology/approach Two preliminary experiments were conducted to identify the types of AM knowledge that need to be transmitted and in what format, as well as the best time to integrate this knowledge into the design process. Accordingly, a six-phase knowledge transmission framework was developed to run in parallel with the design stages in a design for AM (DfAM) approach. This framework was then validated through three additional experiments. Findings The proposed framework enables a more informed design process by providing non-expert designers with relevant AM information. Its effectiveness was validated through experiments involving students with no prior AM knowledge, demonstrating improved design outcomes and reduced need for iterative rework. Originality/value This work stands out by proposing a sequence for providing information that is compatible with existing design methods. It offers a versatile approach that directly connects theoretical design methods with practical guidance, helping designers make more informed and effective decisions when designing for AM.

Additive manufacturing (AM) processes rely heavily on the geometric intricacies of parts being produced. Variations in the geometry of sequentially printed layers can introduce defects, arising from the complex interactions between geometry, process parameters, and materials. To mitigate these errors, it is critical to identify key geometric features and track their evolution in terms of both location and timing throughout the design and manufac- turing stages. This paper presents a novel method to describe shape in the context of AM, emphasizing the importance of layer-wise material deposition. Mereotopology, a framework for qualitatively describing the relationships between parts and wholes, is employed to pro- vide insights into the spatial, temporal, and spatio-temporal relationships inherent in AM processes. This formalism is particularly suited for shapes where precise dimensions may be indeterminate but where relational descriptions can still offer valuable information. The proposed theory integrates spatio-temporal evolution based on mereotopological principles and is applied to benchmark cases in AM. Additionally, a methodology for visualizing the generated descriptions is provided, along with a detailed case study. The paper concludes with a discussion on the strengths and limitations of this theory in comparison to existing approaches for describing four-dimensional objects.

A novel approach for 3D objects clustering to enhance detail design phase is developed with the Hierarchical Clustering Algorithm (HICA). This bigdata analytic extract features like vertex count, genus or convexity. The aim is to classify a 3D objects database and their compatibility with different manufacturing technologies (casting, milling and additive manufacturing), thus facilitating more informed decision-making for designers. The result is the identification of main features based on Variance Inflation Factor (VIF) that enables to evaluate clusters that share similar characteristics. Mean dihedral angles, minimal thickness, betti numbers, accessibility score are found. Then principal components are employed to offer specific information and enable interpretation. This latter can be directly applied to refine and adapt their designs for various manufacturing technologies or use specific segmentation tools.

Wire Arc Additive Manufacturing (WAAM) is an advanced fabrication technique that uses wire as feedstock and an electric arc as a heat source to build metal components layer by layer. Known for its high deposition rates and cost-effectiveness, WAAM is particularly suited for large-scale, complex structures.In this study, a design of experiments was used to investigate the influence of operating parameters on the mechanical and geometric properties of 316L stainless steel walls. Multi-layer and multi-pass (each layer with two U-shaped beads) walls were deposited using the WAAM process with Cold Metal Transfer (CMT) technology. The "Zigzag" strategy, where the endpoint of one layer is the start of the next, was employed to ensure uniform wall height. The substrate used was a 304L stainless steel plate, mounted on an aluminum support to reduce distortions from residual stresses and improve heat dissipation. Machining and micrographic analysis were performed after the additive manufacturing process to evaluate the surface quality and internal structure.The results demonstrated that the operating parameters significantly influenced the geometric precision of the walls. Thermal interactions between layers and within beads played a key role in determining final wall characteristics, and careful parameter control is essential to optimize both geometry and mechanical performance.

The absorptivity is a major parameter in laser-matter interaction. It can be measured by different methods (thermal or optical). The already known integrating sphere method is generally used with spheres whose choice of dimensions and architecture is not discussed in the literature. The influence of the parameters of the sphere (size of the sphere, position of the photodiode on the sphere, distance of the photodiode from the sphere, opening of the passage of the laser, distance from sample to the sphere, presence of a baffle in front of the photodiode, sample angle, coating materials) was tested with a mirror of known reflectivity. This study has demonstrated that it is possible to use small integrating spheres for absorptivity measurements. These spheres are home-made by additive manufacturing from a polymer, with the inner walls coated with BaSO4. The optimum experimental conditions for these small spheres are defined, particularly the angle sample. The absorptivity of various materials in the solid state and powder bed was measured for four wavelengths. These measurements were carried out for three sizes of spheres and confirmed the literature results (for example, the absorptivity is measured at 80% for copper at blue wavelength). This small sphere will be adapted to in-situ measurement, particularly in the Laser Powder Bed Fusion process whatever wavelength, particularly new green or blue laser sources, and material.

Wire-arc additive manufacturing (WAAM) is an advanced technique for fabricating large metal components through layer-by-layer material deposition using arc welding methods. This study focused on optimizing the WAAM process by employing machine learning models to predict and control bead geometries, specifically bead height (BH) and bead width (BW), while ensuring consistent height increments in multibead walls. Based on CMT technology in cold metal transfer experiments, linear regression models achieved high accuracy in predicting BH and BW. Analysis of variance results highlighted the considerable influence of voltage (V) and travel speed (TS) on bead geometries. For multibead wall characteristics, polynomial regression models incorporating non-linear terms, such as travel speed (TS²) and dwell time (Dt²), were developed to predict height (H) and waviness (W). Various optimization metrics were employed to balance the trade-offs between H and W for identifying optimal welding conditions that achieved the target H while minimizing W. A notable innovation of this research is the optimization of dwell time (Dt) for each layer to achieve a linear incremental H profile, minimizing W and ensuring consistent layer quality.

This study proposes a new approach on a multiscale analysis of 316L stainless steel microstructures to enhance the predictability and homogeneity of microstructure for the Wire Arc Additive Manufacturing (WAAM) process. Despite the promise in the fabrication of large-scale metallic components, achieving consistent microstructure and mechanical properties remains a challenge with WAAM. This research investigates the solidification behaviors, grain morphology, and mechanical characteristics of 316L stainless steel, aiming to develop a predictive framework for WAAM process optimization. By employing a systematic approach to the fabrication and subsequent analysis of 316L stainless steel walls, the study reveals critical insights into thermal gradients, solidification rates, and their impacts on microstructural features. The findings are anticipated to inform improved fabrication strategies, leading to enhanced mechanical properties and reliability in WAAM-manufactured components based on mereotopology philosophical framework. A spatio-temporal of the different tracks could be defined based on trajectory analysis.

AbstractResearch on additive manufacturing has highlighted methods and guidelines to optimise the design process and improving finished product quality. There is still room for improvement in making AM as reliable as more traditional processes when considering industrial use. In terms of manufacturing, managing print parameters properly can improve reproducibility and repeatability of a part, in addition to its fidelity to the basic geometric model. However, a topological optimised geometry requires more than good parameterisation. Efforts are therefore being made to formalise knowledge so that it is explicit and accessible to designers. This paper proposes an approach based on the spatio-temporal evolution of a geometry during printing to quantify data at the meso scale. Previous studies have been conducted on the description of features in time, space and space-time, and on the influence of their arrangement within a part. Building on this work, a parameterised test specimen was designed to measure the quantitative impact of these arrangements on the final product. The method is then presented and illustrated through a case study to help the designer with quantitative predictive values of geometric parameters.

The effective absorptivity of IR laser light for different powder beds were studied. The reflectivity of aluminum, titanium, stainless steel and copper alloys was measured using an appropriate Ulbricht sphere. Laser irradiation was reliably detected by a photodiode. Reflectivity was carefully measured as a function of illuminated area and powder bed density. Several powder size distributions and powder thicknesses were chosen to evaluate the impact on the laser absorption. Two spot diameters were tested to evaluate the variation of the reflectivity. The absorptivity of the powder bed was significantly higher than the absorptivity of a uniform surface for similar material due to multiple scattering. In addition, the substrate is responsible for a non-negligible variation in the powder bed absorption. The inhomogeneity of the powder bed strongly modified the laser absorption for a small spot size. The absorption fluctuated during the transition from the powder state to the molten pool state.

Wire arc additive manufacturing is a process with great potential and in full expansion for structural, maintenance, and large-scale product development. It is gaining interest due to its low buy-to-fly ratio. Nevertheless, its major flaw is the low geometric accuracy and the wavy surfaces of the printed parts. To ensure that the requirements of the final part are respected, the dimensions and the minimum quantity of material to be machined have to be specified. Thus, machining allowance and part dimensions are elements that were quantified based on a design of experiments (DOE) and thermal analysis. It was used to determine the effects of cold metal transfer (CMT) parameters on them. For instance, it was found that travel speed and interpass time have a major influence on height, width, and machining allowance. Specific way to achieve the desired requirements is discussed.

AbstractAdditive manufacturing (AM) processes are now integrated in industry. Therefore, new methods to design AM parts taken into consideration capabilities and limitations are necessary. It is very difficult for teachers to effectively guide students with ideas emerging from generative design tools. AM requires significant preparation and compromises. Topological optimization is also used depending on requirements. A significant impact on the final part quality is related to the part orientation and geometric dimensions. Therefore, this white paper focuses on detailed design steps to prepare future technicians and engineers to design for additive manufacturing. Active teaching pedagogy guideline is proposed. Students have to think in 3D and use analysis tools to create and validate the optimised design. They use immersive tools to review constraints and model diagnostic algorithm to generate data. Present approaches with design guidelines and tools enable to create AM rules based on it. Questionnaire shows that students need explicit knowledge information. Features recognition and geometry diagnostic are mandatory for complex model. Immersive tool helps to evaluate post-processing. They can now relate AM product-process relationship.

Opportunities are offered by multiple Additive Manufacturing (AM) processes nowadays. Design rules are evolving to lead to lighter and stiffer parts with really more complex shapes than those obtained by conventional processes. Worldwide, new methodologies/tools of assistance for the design are developed such as Design for Additive Manufacturing (DfAM). Additive manufacturing can allow the development of new metamaterials and health-matter evaluation based on energy flow evaluation. In this paper, the objective is to generate a new methodology with DfAM based on mesoscale knowledge. It is generated with open lab bench and simple object characterization. A methodology is presented to formalize and quantify information at multilayer dimension. A database is also generated following Design of Experiments (DoE) to obtain metamodels. They are developed for specific features representative of AM geometric class such as overhanging, holes or walls for instance. Mereotopological primitives with their AM definitions are used to define features in term of space and time variables. This theory enables the formalization of knowledge at the mesoscopic scale taken into consideration layer by layer build-up. It is then possible to use it to integrate data and information to the different feature juxtapositions using recognition algorithm. Information for each feature can then be included and explicitly used to help the designer during detailed design phase. A global 4-steps DfAM methodology maximizing the potential of AM is presented and validated through a part from the space industry use case. It includes the definition of skeleton/skin entities, pattern decomposition, information associated based on material evaluation and decision for AM part.

In the context of the Industry 4.0, new processes have appeared, such as the additive manufacturing (AM) process. Therefore, new approaches to design parts have to be developed to integrate process constraints. It is very difficult for teachers to effectively guide students during conceptual design for AM, even though various idea generation techniques and methods are available. AM requires an important preparation and compromise in design phases. In addition, design need to be generated in a digital environment. Among the various steps, critical impacts on the final part quality are linked to part orientation. So, this paper focuses on the conceptual design phase to educate future technician and engineers to the design for additive manufacturing. Pilot-study on the teacher's role interacts through active pedagogical tool with students. They need to think in 3D and create directly in 3D. The propose education development use an immersive tool to consider the process constraints. Thereby, students need to deal with an AM process chain. New approaches are analyzed based on the design guidelines for Additive Manufacturing, which were developed by the students themselves. Also, the students estimated opportunities and limits linked to product-process relationship. Finally, the success of the new course contents and form is reviewed by a student evaluation.

Additive Manufacturing (AM) has many advantages, but the lack of access to the knowledge associated with it minimises its development in industry. The design phase is crucial for the success of AM, a challenge for Design for Additive Manufacturing (DfAM) methods is therefore to facilitate the access and manipulation of this knowledge. This transfer of knowledge can be achieved by formalising rules at all scales, and communicating them to the designer at the appropriate phase. It is hence necessary to find a way to formalise information in time, space and space-time dimensions since AM is a process that places material in space and layer by layer. The concept of mereotopology is used to study the relationships of connection and interaction between parts, wholes and boundaries, and may be a suitable resource to study DfAM along these three dimensions. The aim of this paper will therefore be to present a method for searching and formulating design guidelines based on a discretisation of the process enabled by the concept of mereotopology. This method consists in the decomposition of a 3D model into features between which spatial, temporal and spatio-temporal interactions are studied. Simultaneously, the analysis of manufactured defects on a printed version of the model allows to link manufacturing defects with a configuration of spatial and temporal elements. Once the defects and configurations have been linked, rules are formulated and then validated or invalidated according to their recurrence on different models printed with the same process and material. This method could be integrated in industry to take advantage of manufacturing defects in order to add data to the statistical study.

Additive Manufacturing are x² evolving with data monitoring possibilities in industrial machine. This work is focused on tools for the development of health matter evaluation. It is a first step before developing in-situ control strategies. One process is studied: Wire Arc Additive Manufacturing process. The work presents the design specification necessary of a lab-bench and a methodology with in-situ analysis to generate defect detection metamodel. The methodology for metamodel is presented in this article. Data are obtained from a photodiodes and machine learning algorithms are used to generate two metamodels, after the definition of a design of experiment using a SOBOL tool, on the average and the standard deviation. The influence of the travelling speed and the current intensity are studied, on the sensor signal and the samples quality, to illustrate the potential of in-situ monitoring for the process. The accuracy of the prediction is also presented, like the sensitivity of the parameters on the results.

Wire arc additive manufacturing process (WAAM) is an innovative technology that offers freedom in terms of designing functional parts, due to its ability to manufacture large and complex workpieces with a high rate of deposition. This technology is a metal AM process using an electric arc heat source. The parts manufactured are affected by thermal residual stresses due to high-energy input between wire and workpiece despite numerous advantages with this technology. It could cause severe deformation and change the global mechanical response. A 3D transient thermal model was created to evaluate the thermal gradients and fields during metal deposition. The material used in this study is a steel alloy (S355JR-AR). This numerical model takes into account the heat dissipation through the external environment and the heat loss through the cooling system under the base plate. Birth-element activation strategy was used to generate warm solid part following the movement of the heat source. The metal deposition is defined with constant welding speed. Goldak model was used to simulate the heat source in order to have a realistic heat flow distribution. Results were in concordance for thermal cycles at different points comparing with experimental results issued from bibliography in terms of: (1) Temperature maximum, (2) Thermal cycles and (3) Cooling gradient phase. This study enabled to check the numerical model and used as a predictive tool

Nowadays products extend their capabilities towards changing their configurations in order to cover multiple usage needs. They may be named transformable products and have not been taken into consideration in early design stages yet. In this paper, a proactive definition of the product is provided with transformation intrinsic properties. The formalization leads to an architecture. This enables developing a transformable product from two ordinary non-evolving objects. Different configurations and transformation processes have been set and implemented within a CAD tool to design a transformable product. A new paradigm is thus initiated, which will lead to efficient and dynamic design of transformable product.