TOMRA Recycling, a global leader in sensor-based sorting technologies, has developed GAINnext™, an innovative solution based on artificial intelligence and Deep Learning to improve the recovery and purity of forged aluminum in the metallurgical industry. This technology enables highly accurate identification and separation of forged aluminum scrap from low-alloy casting, optimizing the quality of recycled material and reducing impurities such as silicon.
GAINnext™ complements TOMRA’s X-TRACT™ technology, which uses X-rays to sort metals by atomic density. While X-TRACT™ separates heavy metals and produces high-purity aluminum waste, GAINnext™ uses RGB cameras and neural networks to analyze thousands of images per millisecond, distinguishing materials by shape and size with speed and precision superior to human vision.
In addition, the purity of the forged fractions obtained can be further improved with AUTOSORT™ PULSE, a system that uses laser spectroscopy to accurately identify and classify different types of alloy.
Tom Jansen, Sales Director for the Metallurgical Segment at TOMRA, emphasizes that this technology expands the capabilities of the GAINnext™ ecosystem, offering an efficient and economical solution that reduces the need for manual separation and allows recyclers to produce high-quality recycled aluminum for a circular model.
Since 2019, TOMRA has applied Deep Learning in various areas of recycling and will continue to expand these technologies with new applications in 2025.