This innovative solution integrates physics-informed probabilistic machine learning with PET blow molding processes to enhance design accuracy and efficiency while addressing material and process uncertainties.
The AI-powered automated design system for PET containers leverages advanced machine learning techniques to optimize the blow molding process. By integrating physics-informed probabilistic models, this solution enhances the accuracy, robustness, and adaptability of automated container design. The approach addresses material and process uncertainties, significantly streamlining the design process and setting new industry standards.
Key features:
Currently at Technology Readiness Level 3, this solution has undergone initial validation stages with a focus on model development and integration, preparing for further data-driven refinements and real-world applications.
The University of Washington is a large public research university with campuses in Seattle, Bothell, and Tacoma, known for a broad portfolio from fundamental discovery to applied innovation. Industry partners engage through a South Lake Union research campus adjacent to a major life sciences district and through collaboration programs that place faculty and students alongside corporate R&D. The university’s integration with a major academic health system enables clinical translation and large-scale trials. Research is supported by competitive federal funding from NIH, NSF, DOE, and DoD. A dedicated technology transfer office manages IP, licensing, and startup incubation with prototyping resources.