Artificial Intelligence and Machine Learning Approaches in Eco-Design for Additive Manufacturing: A Literature Review

Additive manufacturing (AM) is currently considered one of the most promising technologies for fostering innovation in the industrial sector, specifically in product design. Considering the non-subtractive nature of AM processes, in first approximation, they all seem to be environmentally sustainable since they lead to a null or at least reduced quantity of scraps. However, the peculiarities of such technologies, such as the specific input materials (e.g., metal powders, polymer filaments) or the low productivity, make the situation more complex, not so homogeneous in all the cases, and strongly dependent on the process and product parameters. In this complex scenario, the present paper aims to investigate the current state of the art regarding adopting Artificial Intelligence (AI) and Machine Learning (ML) toward the eco-design for AM. Following the PRISMA methodology, a set of 19 scientific papers were analysed. Papers were grouped according to the benefits (improvement of quality and reduction of energy, waste and defects) and design opportunities (definition of shapes and selection of materials) that AI and ML potentially provide. The discussion of the results presents the most relevant research and development opportunities for scholars and companies.