Manufacturing
Overview
Manufacturing research uses data, models, and automation to improve how products are designed and made, from CAD-based process planning to shop-floor scheduling and robot coordination. Common approaches include deep learning for inspection and sensor analysis, reinforcement learning for scheduling and process control, physics-informed models for predicting process behavior, and knowledge graphs or language-model agents for connecting instructions, machine data, and decisions. Researchers evaluate these methods using production records, simulators, digital twins, and benchmarks for tasks such as defect detection, machine understanding, and cross-machine transfer. A central challenge is making systems reliable beyond controlled tests despite scarce or shifting data, while keeping decisions physically feasible, interpretable, and safe for human-supervised production.
Latest papers
77 papers10 papers in the last four weeks, up 25% on the four weeks before. 0.1% of all new papers.