FMRG: Cyber: Custom Product Manufacture at Metal Recycling Plants Using Artificial Intelligence, Extended Reality, and Human-Robot Collaboration

About This Grant

This Future Manufacturing Research Grant (FMRG) project funds research that attempts to make novel strides in Future Manufacturing by contributing new cyber technologies to enable recyclofacturing, or custom manufacturing of products to user specifications using scrap or consumer material at metal recycling facilities. The project looks to contribute advances in artificial intelligence, extended reality, and human-robot collaboration to transform metal recycling facilities, that focus on traditionally low-skilled tasks, into new and agile product manufacturers, helping to keep manufacturing in the US. The project’s contributions could transform the expertise-heavy workflow of traditional computer-aided design and manufacturing into an intuitive workflow that enables recycling plant workers to translate an end user's arbitrary product specification into a product build, upon rapid upskilling to the novel cyber technologies. The education and workforce development plan seeks to improve the job outlook for current and future recycling plant workers by upskilling them to become new-collar workers through courses in prompt engineering for recyclofactured product design, extended reality-based weld training, and working with collaborative robots for assembly of recyclofactured products. The project seeks to transform recycling facilities into agile manufacturers attractive to future workers who increasingly look toward technology-enhanced trades. The team is partnering with 2- and 4-year colleges providing vocational training and workforce development, business development agencies, and groups supporting individuals in manufacturing and welding to conduct interviews, perform course development, and conduct evaluations. The project provides benefits to society by strengthening community/recycler networks to spur local product fabrication and enhancing self-sufficiency of local communities. The research looks to make fundamental contributions in three areas. First, it seeks to advance artificial intelligence by providing neural networks for automated generalized computer-aided design model generation from user product specifications. These models are anticipated to include text, images, and sketches, automated scrap metal sheet assessments, optimized model decomposition and facet assignment to sheets, and next step build predictions. Second, it looks to provide novel extended reality work for guidance to the human worker in the build process through overlay of expert recommendations and next step structure, enabling enhanced human-in-the-loop involvement. Third, it looks to develop human-robot collaboration for assembly of recyclofactured products with dynamic task allocation between humans and collaborative robots by integrating awareness of worker load, preferences, activity, and ergonomic risk. The Future of Work assessment integrates qualitative interviews with workers and managers, large-scale societal survey of expectations from cyber-enabled manufacturing, quantitative assessment of demand and production costs through willingness-to-pay experiments and labor cost estimation by connecting with research plan studies on technology interaction, and projection models to inform on long-term impact of recyclofacturing. This FMRG award was supported by three Engineering Divisions including: Civil, Mechanical, and Manufacturing Innovation (CMMI); Engineering Education and Centers (EEC); and Chemical, Bioengineering, Environmental, and Transport Systems (CBET). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.