Due to the great freedom of design, additive manufacturing (AM) offers enormous potential for lightweight construction. In order to harness this potential, it is crucial to integrate knowledge between the individual stages of the AM value chain. By digitally linking AM process and product data, lightweight design and production can be further optimized in economic and ecological terms. The present work demonstrates the added value for the lightweight potential of additively manufactured (laser powder bed fusion (LPBF)) aluminum components when decentralized material and AM process data is shared between actors via a cross-institutional data space.

The use case discussed here describes a design engineer who optimizes a component originally manufactured by casting AlSi10Mg with regard to lightweight design by means of topology optimization for AM. A process-specific topology optimization algorithm (PSTO) provides an optimized design and also knowledge about the best suited combination of AM-machine, part orientation in build space and a heat treatment. As input for this design algorithm, data about mechanical material characteristics, AM process parameters and post-processing information such as heat or surface treatment is required.

Since different data is often provided by different actors, compliance with the FAIR data principles [1] is paramount. To meet this need, a data space architecture based on the International Data Spaces (IDS) reference architecture model is designed and implemented. By doing so the authors create the very first instance of the Materials Data Space® (MDS). Using the MDS in this context ensures findability (F) of the data via a, safe access (A) to the data from the company’s IT-infrastructures, interoperable (I) data exchange under highest data sovereignty standards and reusability (R) of the data by the PSTO due to rich metadata in form of knowledge graphs. The knowledge graphs describe the entire process chain from additive manufacturing to mechanical testing and embed the resulting data in its context through an ontology-based process model. These semantic metadata enable the linking of data in the sense of a cross-institutional data fabric with the aim of automated data analysis for an intelligent lightweight design by the designer engineer.

The PSTO uses adapted material and process-specific modeling depending on the data obtained from the MDS in a multimodal optimization scheme. When using the decentralized data from the MDS, the optimized design results in a mass reduction of 20 % in comparison to the cast part and 67 % reduction of detrimental support structures in LPBF, which is a significant improvement with respect to the state-of-the-art topology optimization algorithms without linkage of material and process data.

[1] Wilkinson, M. D. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci. Data 3:160018 doi: 10.1038/sdata.2016.18 (2016)