2026 Roadmap on Artificial Intelligence and Machine Learning for Smart Manufacturing
Covers 30+ topics from digital twins to LLMs in a single comprehensive guide.
A consortium of 54 researchers led by Jay Lee (University of Cincinnati) has published the '2026 Roadmap on Artificial Intelligence and Machine Learning for Smart Manufacturing' in the journal Machine Learning: Engineering. The comprehensive roadmap is structured in three parts: first, it lays out the foundational trends shaping AI evolution in manufacturing. Second, it details key active areas where AI already drives advances—including industrial big data analytics, advanced sensing, autonomous systems, additive and laser-based manufacturing, digital twins, robotics, supply chain/logistics optimization, and sustainable manufacturing.
The third section explores non-traditional ML approaches opening new frontiers: physics-informed AI, generative AI, semantic AI, explainable AI, RAMS (reliability, availability, maintainability, safety), data-centric metrology, and large language models (LLMs) and foundation models for highly connected manufacturing systems. The roadmap also highlights persistent challenges: industrial big data complexity, effective data management, integration with heterogeneous sensing/control systems, and the demand for trustworthy, explainable, reliable operation in high-stakes environments. It aims to guide researchers, engineers, and practitioners in accelerating innovation and aligning academic and industrial priorities for scalable, sustainable AI-driven manufacturing.
- Covers 30+ subtopics from digital twins to LLMs and physics-informed AI
- Identifies persistent barriers: industrial data complexity, trustworthiness, integration with legacy systems
- Accepted in the peer-reviewed journal Machine Learning: Engineering (DOI: 10.1088/3049-4761/ae5967)
Why It Matters
This roadmap aligns R&D priorities for the next wave of autonomous, sustainable, and AI-driven factories.