Use AI and multi-objective algorithms to optimize PET molding processes that incorporate post-consumer recycled (PCR) resins. The approach predicts and minimizes cycle-time variations to support consistent production quality across different designs and machine configurations. It combines machine learning with decision-support visualization to balance cycle-time reduction against quality maintenance.
This solution uses artificial intelligence and multi-objective evolutionary algorithms (MOEAs) to improve the efficiency of PET molding processes that incorporate post-consumer recycled (PCR) resins. It predicts and minimizes the impact of PCR content on injection molding cycle times, aiming to support uniform manufacturing efficiency and product quality. The predictive model is designed to adapt to different preform designs and machine configurations, targeting versatility for real-world manufacturing environments.
The technology is at TRL 2, indicating that the concept and application have been formulated. Ongoing validation includes data collection, model development, and system integration to support further refinement and testing.
The Catholic University of America is a private national research university in Washington, D.C., known for a collaborative, mission-driven culture and comprehensive academic programs. For industry partners, its location places teams within minutes of federal agencies, standards bodies, and policy organizations, enabling early regulatory input and access to decision‑makers. Companies engage on-campus core facilities and sponsored-research labs for contract R&D, prototyping, and testing, and connect with talent through internships and project-based collaborations. Research is supported by competitive federal funding from agencies such as the National Science Foundation, National Institutes of Health, the Department of Defense, and the Department of Energy. A dedicated technology transfer office manages IP, licensing, and startup formation with industry-friendly agreements.