Precision Perfected: How AI/ML Drives Next-Generation Manufacturing

Artificial Intelligence (AI) and Machine Learning (ML) hold significant potential for driving efficiency, automation, and new capabilities within the manufacturing industry. The key impediment to achieving this goal includes a lack of large-scale data available to develop models (often due to privacy concerns), shortage of experts that know both AI/ML and manufacturing, and the substantial upfront investment required for infrastructure and integration.

This talk will provide an overview of efforts at Carnegie Mellon University (CMU) to apply AI and ML technologies to advanced manufacturing and to develop the next-generation workforce for this challenge. We are creating testbeds for advanced manufacturing and a digital backbone to collect, collate, and contextualize this data. Technologies emerging from this effort include CNN-based algorithms for real-time defect detection (on the edge), the use of LLMs for manufacturing process control (without pre-training), and the application of AI in computer-aided manufacturing.

A specific problem addressing process drifts in jetting-based electronics printing, known as Aerosol Jet 3D printing, will then be detailed. I will conclude by discussing the prospects in this area.