A Review on Machine Learning, Big Data Analytics, and Design for Additive Manufacturing for Aerospace Applications

Additive manufacturing (AM) has emerged as a promising technology to cater to the increasing demand for the fabrication of multi-functional, multi-material, and complex parts. AM is revolutionizing production and product development in the aerospace, automotive, and medical fields. However, mismatch in material properties, pervasive imperfections in the printed part, and lack of build consistency are crucial concerns. Higher accuracy in AM processes primarily depends on controlling various aspects of the process. In the last few years, machine learning, data analytics, and design for additive manufacturing have been the most extensively used techniques to address the vital concerns of additive manufacturing. Despite well-known techniques, very few studies reported applications of these techniques for aerospace. Specifically, this study comprehensively reviews recent advancements in the design for additive manufacturing (DfAM) and applications of machine learning and big data analytics to address the prime concerns of AM. The DfAM emphasizes issues and opportunities for topology optimization and methods for generative design for weight reduction and manufacturing of products with high resolution. Simulation and modeling techniques that are being used to improve geometric quality and process analysis are discussed to enable its potential for different applications. Further, automation of AM process using the Internet of things and knowledge-based systematic process planning is discussed to address key issues in process planning of multiple parts. Finally, the current challenges and scope for algorithmically driven AM processes are summarized with the trends of automation in AM to ensure greater efficiency and a better lifecycle of AM products in the era of industry 4.0.

Automated summary

Additive manufacturing (AM) is a promising technology for fabricating multi-functional, multi-material, and complex parts. This study comprehensively reviews recent advancements in design for AM and the applications of machine learning and big data analytics to address the concerns of AM processes. The study highlights topology optimization, generative design, simulation and modeling techniques, as well as automation and knowledge-based process planning in AM. The challenges and trends in algorithmically driven AM processes are also discussed.