SBIR Phase II: Simplifying the use of recycled plastics in film extrusion

Award ID: 2212917

The broader impact of this Small Business Innovation Research (SBIR) Phase II project will be to allow recycled plastics to be used more efficiently and affordably than is currently possible. Regulatory and societal pressures are forcing reconsideration of single use plastics, and manufacturers of plastic film must use recycled plastic at higher quantities. However, it is difficult for the manufacturers to affordably reincorporate single-use plastics due to the low quality and unpredictable content of the material. Increasing yield of usable plastics through use of the proposed technology is expected to reduce waste, offering the potential to annually save 6.5 million metric tons of carbon dioxide emissions in the US and Canada, and 28 million metric tons globally. Also, the greater use of artificial intelligence in manufacturing is of strategic advantage to the US, with the proposed technology also applicable to metals, paper, or advanced materials. Furthermore, skills shortages are impacting manufacturing and are likely to worsen due to a rapidly aging workforce. A great deal of on-the-job expertise will be lost in the coming years as a generation of experienced operators retires. The proposed solution can ease this transition, acting as an expert decision system to carry the intelligence forward and help maintain US manufacturing competitiveness. This Small Business Innovation Research (SBIR) Phase II project will apply artificial intelligence (AI) capabilities and process control methods to plastic film extrusion, and subsequently to other types of manufacturing. Currently hardware solutions exist for manufacturers, though they can be expensive, difficult to use and maintain, and can require specialized skills to use. By contrast, the proposed technology is a software-based approach to the control of complex plastic film extrusion processes, particularly in the context of widely variable input materials such as recycled plastics. The AI software will be robust to changes in the production environment and will account for process drift over time. These technology capabilities are industrially novel and not known in the academic literature. Phase I outcomes suggest that the technology can automatically control extrusion processes to achieve optimal steady state production faster than is the currently possible via human control. The AI-based expert system effectively recreates the knowledge tacitly held by long-experienced factory operators. This type of industrial automation has the potential to be value-generating for the wider manufacturing sector. The proposed technology may be applicable to a wider range of extrusion manufacturing processes, such as extrusion of metals, paper or advanced materials. 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.

NSF Program Director: Peter Atherton