AI in Manufacturing

Artificial intelligence in manufacturing is not new.

Manufacturing has been using Artificial intelligence (AI) since the deployment of expert systems, a type of AI, in the 1960s (Xu et al 2022). Current AI technological advancements include machine learning (ML), deep learning, and large language models (LLMs).

Current AI technology has the potential to help manufacturing in predictive maintenance, quality assurance, and process monitoring and optimization (Plathottam et al 2023). This is due in part to the large quantities of available data from processes, sensors, and equipment that can be analyzed using ML and other AI (Arinez et al 2020). However, AI can pose risks, including the trust, implementation, infrastructure, and human resources challenges common to new technologies in manufacturing (Plathottam et al 2023). Similarly, identifying appropriate use-case scenarios and algorithmic tools is a barrier to implementation (Nti et al 2021). In addition, AI brings data acquisition, management, and security challenges.

Human, technological, and operational factors determine successful AI adoption.

Research-intensive, knowledge-based, and service-oriented companies tend to be the first firms to adopt AI technologies (Kinkel et al 2022). These firms implement AI technologies at both domestic and foreign production sites at the same rates.

Deployment criteria for AI in manufacturing mirrors AI deployment in other sectors (see the Science Note, State Regulation of AI & Workforce). Broadly, success in manufacturing depends on three categories (Kutz et al 2022):

• Human – Maximizing human participation, encouraging user engagement and feedback, managing expectations, fostering trust and cooperation, removing barriers between IT and other teams, and training workers
• Technical – Utilizing appropriate datasets, standardizing software and systems, ensuring user and data privacy, creating appropriate algorithmic transparency, reducing algorithmic bias, and validating before large-scale implementation
• Operational – Having a clear value-add for AI, establishing clear communication and fostering a culture of transparency, keeping a holistic view of projects to synthesize solutions across departments and teams, ensuring appropriate data safety and governance, creating rapid feedback and development loops, defining roles, responsibilities, and processes, and focusing on outcomes beyond profitability

Workers need continuous training.

Research suggests that manufacturing firms will need to invest in and encourage continuous learning and career development as part of organization-wide strategic goals (Li 2022). This investment is easier for large and medium-sized firms, who are able to train their entire workforce at once, while smaller firms train on a need-to-know basis (Pedota et al 2023). Firm culture and commitment to continuous learning encourage professional development among the workforce (Leon 2023).

Manufacturing covers and requires a wide range of skills, and technical competencies are constantly changing. Research estimated that 67% of skills considered important in 2020 would change by 2025, and that a third of skills required in 2025 will be in competencies that were not considered crucial in 2020 (Li 2022).

To address this skill gap, Learning Factories (LFs) are a key tool used to upskill and retrain existing workforce on new technologies (Dehbozorgi et al 2024, Gyulai et al 2023). LFs are a realistic manufacturing environment where education, training, and research occurs (Purdue University 2025, Neacsu et al 2021, Figure 1). LFs bring together students, practitioners, and educators, also known as communities of practice, for informal and formal learning, creating a connection between industry and academia (US EDA 2025, Neacsu et al 2021). LFs enhance training quality, facilitate technology transfer, develop interdisciplinary competencies, and enhance workers’ soft skills, motivation, and overall work attitude. Communities of practice have been identified as some of the most successful programs for re- and up-skilling (Leon 2023). Similarly, programs where managers, educators, and students are involved in developing curriculum and identifying skill gaps have shown promise for appropriately reskilling workers (Geraldes et al 2021).

Figure 1. Learning Factories (LF) models. Stand-alone (top) versus industry partnered projects (bottom). Figure from Neacsu et al 2021.

References

Arinez JF, Change Q, Gao RX, Xu C, Zhang J (2020) Artificial Intelligence in Advanced Manufacturing: Current Status and Future Outlook. Journal of Manufacturing Science and Engineering, 142(11): 110804. https://doi.org/10.1115/1.4047855

Dehbozorgi MH, Rossi M, Terzi S, Carminati L, Sala R, Magni F (2024) AI Education for Tomorrow’s Workforce: Leveraging Learning Factories for AI Education and Workforce Preparedness. 2024 IEEE Forum on Research and Technologies for Society and Industry Innovation (RTSI), 677-682. https://ieeexplore.ieee.org/abstract/document/10761217

Geraldes CAS, Fernandes FP, Sakurada L, Rasmussen AL, Bennyson R, Pellegri U (2021) Co-Design Process for Upskilling the Workforce in the Factories of the Future. IECON 2021 – 47^th Annual Conference of the IEEE Industrial Electronics Society. https://ieeexplore.ieee.org/abstract/document/9589528

Gyulai T, Viharos ZJ, Kasa F, Wolf P (2023) Upskilling SME Workforce by Learning Factories. Proceedings of the 13^th Conference on Learning Factories (CLF 2023). https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4469802

Kinkel S, Baumgartner M, Cherubini E (2022) Prerequisites for the adoption of AI technologies in manufacturing – Evidence from a worldwide sample of manufacturing companies. Technovation. 110. https://doi.org/10.1016/j.technovation.2021.102375

Kutz J, Neuhuttler J, Spilski J, Lachmann T (2022) Implementation of AI Technologies in manufacturing – success factors and challenges. Conference Paper. The Human Side of Service Engineering, 62: 256-261. https://doi.org/10.54941/ahfe1002565

Leon RD (2023) Employees’ reskilling and upskilling for industry 5.0: Selecting the best professional development programmes. Technology in Society, 75. https://doi.org/10.1016/j.techsoc.2023.102393

Li L (2022) Reskilling and Upskilling the Future-ready Workforce for Industry 4.0 and Beyond. Information Systems Frontiers, 26: 1697-1712. https://doi.org/10.1007/s10796-022-10308-y

Neacsu GC, Pascu IG, Nitu EL, Gavriluta AC (2021) Brief review of methods and techniques used in Learning Factories in the context of Industry 4.0. IOP Conference Series: Materials Science and Engineering. 11^th International Conference on Advanced Manufacturing Technologies, 1018. https://iopscience.iop.org/article/10.1088/1757-899X/1018/1/012022/meta

Nti IK, Adekoya AF, Weyori BA, Nyarko-Boateng O (2021) Applications of artificial intelligence in engineering and manufacturing: a systemic review. Journal of Intelligence Manufacturing, 33: 1581-1601. https://doi.org/10.1007/s10845-021-01771-6

Pedota M, Grilli L, Piscitello L (2023) Technology adoption and upskilling in the wake of Industry 4.0. Technological Forecasting and Social Change, 187. https://doi.org/10.1016/j.techfore.2022.122085

Plathottam SJ, Rzonca A, Lakhnori R, Oloeje CO (2023) A review of artificial intelligence applications in manufacturing operations. Journal of Advanced Manufacturing and Processing, 5(3): e10159. https://doi.org/10.1002/amp2.10159

Purdue University (2025) Smart Manufacturing Ecosystem. School of Engineering Technology. https://polytechnic.purdue.edu/smart-manufacturing-ecosystem

US Economic Development Administration (EDA) Manufacturing Community of Practice. https://www.eda.gov/strategic-initiatives/communities-of-practice/manufacturing-community-of-practice

Xu J, Kovatsch M, Mattern D, Mazza F, Harasic M, Paschke A, Lucia S (2022) A Review on AI for Smart Manufacturing: Deep Learning Challenges and Solutions. Applied Science, 12(16). https://doi.org/10.3390/app12168239