TOMRA Recycling has introduced a new packaging sorting application for GAINnext, which leverages deep learning – a subset of artificial intelligence (AI) and machine learning – to remove hard-to-classify materials. Further expanding its deep-learning-based applications, TOMRA offers a high-throughput solution for used beverage can (UBC) aluminum recovery that delivers 98 percent purity or higher without manual sorting. This technology automates the sorting line to improve UBC capture efficiency for the material recovery facility (MRF), increasing revenue and decreasing costs.
TOMRA's deep learning solution enables MRFs to maximize the recovery and purity of aluminum from metal packaging waste streams. GAINnext utilizes AI to detect and eject non-UBC materials like aluminum bottles, food cans, trays, UBC metals or plastics, and more, for high-accuracy, automated sorting of aluminum cans. The solution features automated sorting at high belt speeds to improve operational efficiency with up to 33 times more throughput than manual sorting.
Developed as an end-of-line solution for MRFs, GAINnext quickly integrates into existing lines to lower overall costs and improve return on investment (ROI). GAINnext uses an RGB camera, trained by thousands of images, to recognize UBC based on shape, size, dimension and more. Its high-throughput processing delivers up to 2,000 ejections per minute, and the deep learning software identifies overlapping objects and calculates positioning for high-precision, above 98 percent purity sorting. Offering exceptional purity levels, the GAINnext UBC application gives the market an automated process for aluminum can-to-can recycling.
"MRFs typically rely on manual sorters at the end of the line to pick UBC from the metal packaging waste stream," explains Ty Rhoad, TOMRA Recycling's vice president of sales for the Americas. "Manual sorting averages approximately 60 picks per minute, but our highly effective GAINnext AI sorting application offers up to 33 times more throughput. Offering high purity, GAINnext is proven to reduce operating costs and increase revenue and productivity, resulting in a quick ROI."
"TOMRA's experience with AI spans decades, as our optical sorting equipment leverages traditional AI to automate sorting lines," adds Indrajeed Prasad, product manager of deep learning at TOMRA Recycling. "GAINnext is trained to see what the human eye can see and detects thousands of objects by visual differences in milliseconds. The deep learning subset of AI creates a hierarchical level of artificial neurons to solve the most complex sorting tasks. We are delighted that our new application focuses on the critical recovery of UBC aluminum cans and offers customers above 98 percent purity rates."
TOMRA's lineup of AI technologies
TOMRA introduced deep learning AI technology in 2019 with its solution to identify and remove polyethylene (PE) silicone cartridges from PE streams. A second deep learning application focused on wood chip classification, sorting solid wood from wood-based materials like chipboard, plywood, and MDF into individual fractions.
Earlier this year, TOMRA announced five new plastics and paper deep learning sorting applications using GAINnext initially for the European market. Three applications separate food-grade from non-food-grade PET, PP, and HDPE at high throughput rates with purity levels reaching 95 percent. Two non-food applications for the GAINnext ecosystem include a PET cleaner application delivering higher purity PET bottle streams and an application for deinking paper for cleaner paper streams. The new UBC application is TOMRA's region-specific GAINnext sorting application initially targeting recyclers in the Americas.
