Auburn Researchers Use AI and Digital Twins to Improve Manufacturing Operations
Researchers at Auburn University are developing artificial intelligence technologies to detect and prevent assembly-line errors, potentially improving manufacturing efficiency and product quality. Supported by a $500,000 National Science Foundation (NSF) grant, the project combines AI, digital twins, and cyber-physical systems to create more adaptive manufacturing environments.
The project, titled Closed-Loop Digital Twins for Adaptive Human-AI Collaboration in Cyber-Physical Systems, is led by Sathya Aakur, associate professor of computer science and software engineering, with Konstantinos Mykoniatis, associate professor of industrial and systems engineering, serving as co-principal investigator.
The research addresses a common manufacturing challenge: mistakes made during assembly can go undetected until products have advanced through several production stages. The team is developing AI models capable of monitoring manufacturing activities, identifying deviations from expected processes, and determining whether those deviations could cause problems later in production.
At Auburn's Tiger Motors Lab, cameras will capture assembly-line activities while AI systems analyze the footage and update digital twins representing the physical manufacturing process. These virtual models could allow supervisors to evaluate corrective actions before making changes to actual production operations.
An important aspect of the research involves distinguishing between acceptable variations in assembly procedures and mistakes that could negatively affect subsequent tasks. By understanding these relationships, the technology could help prevent errors rather than simply identifying defects after they occur.
The project highlights the growing importance of AI-enabled cyber-physical systems in advanced manufacturing, where intelligent monitoring, computational modeling, and human decision-making work together to improve physical operations. The researchers hope their approach will reduce production disruptions, improve quality control, and support more effective human-AI collaboration.
Source: This article is based on information originally published by Auburn University's Samuel Ginn College of Engineering on October 6, 2026.
Read the original article: https://eng.auburn.edu/news/2026/10/collaborative-research-using-ai-to-optimize-assembly-line-production.html