TOMRA unveils a deep learning-based sorting solution.

With the goal of producing smelter-ready aluminum feedstock from mixed scrap, TOMRA has developed a deep learning-based sorting system that rapidly analyzes and classifies different aluminum types with high precision – resolving a longstanding challenge in industrial recycling and setting a new benchmark for purity in the metals sector.

Faced with the higher costs of materials recovery, cities and companies often landfill aluminum scrap –even when separated – due to outdated sorting systems that, in some cases, are still reliant on manual labor and struggle to consistently recognize and differentiate between materials.

Often arriving as a mixture of alloys – some containing impurities that require additional refinement and limit downstream applications – recycled aluminum can vary widely in market value, with profitability further influenced by tariffs meant to support domestic producers.

TOMRA sorting has historically offered a wide range of high-capacity sensor-based systems to meet the requirements of diverse mining applications, especially when it comes to sorting and upgrading diamond, copper, gold, and other ore

Many mining operations choose the sorting precision of TOMRA tech to selectively increase mill throughput while maintaining enhanced overall efficiency.

Now turning that sorting expertise toward recycling, TOMRA has developed GAINnext –a smart system that uses artificial intelligence to distinguish high-quality wrought aluminum from lower-value cast alloys, a crucial step in producing cleaner, more reusable scrap with less waste.

"Deep learning is a powerful addition to our traditional sensor-based technology, and expanding our GAINnext ecosystem aligns with our broader AI strategy," said Tom Jansen, head of the metals segment sales department at TOMRA Recycling. "Building on our proven success and the significant results our customers have achieved with GAINnext in the waste segment, we are confident in delivering similar value to our metals customers."

Smarter sorting with AI

While AI has been applied in other recycling streams, this launch marks the first application of GAINnext in the metals industry. By leveraging this exacting automation tech, TOMRA aims to optimize inexpensive, premium-grade aluminum recovery.

GAINnext is powered by artificial neural networks and deep learning algorithms, allowing it to analyze materials with remarkable speed and precision. Using RGB cameras, it can process hundreds of thousands of images per millisecond, enabling it to classify materials based on shape, size, and dimension – the equivalent of human visual recognition but at an exponentially higher speed, removing up to 2,000 unwanted pieces per minute.

This enables recyclers to achieve consistent outputs with minimal manual intervention, saving time and money while producing higher-quality results with less waste.

"This latest application enhances dry mechanical sorting of mixed metals, offering a cost-effective solution with a low cost per ton thanks to its high throughput and significantly reduced manual sorting requirements," said Jansen. "Our customers will be able to consistently produce smelter-ready, premium-grade aluminum with exceptional purity levels. This is a significant enhancement to our metal recycling solutions portfolio, and we anticipate wide adoption by recyclers seeking to refine scrap and generate high-quality feedstock for circular manufacturing."

TOMRA

TOMRA sorting has historically offered a wide range of high-capacity sensor-based systems to meet the requirements of diverse mining applications, especially when it comes to sorting and upgrading diamond, copper, gold, and other ore.

Integration with existing tech

TOMRA's X-TRACT technology has long been the industry standard for sorting aluminum from heavy metals based on atomic density using X-ray technology. It processes shredded non-ferrous metals to sort high-purity aluminum from mixed metals but struggles with separating low-alloy cast from wrought aluminum. By adding GAINnext to the sorting process, TOMRA ensures a pure final wrought aluminum scrap ready for smelting.

For recyclers who need to refine the process even further, TOMRA offers AUTOSORT PULSE, a laser-based system that distinguishes between aluminum alloys with enough precision to meet the exacting requirements of different industries.

TOMRA has been developing AI-driven sorting technology since 2019, using it to improve plastic, paper, metal, and wood recycling. In 2024, they introduced an AI system to separate food-safe plastics from non-food plastics, and now are taking metal recycling by storm.

And with more AI-powered solutions already in development for 2025, TOMRA is signaling a continued push to make smarter, faster, and more sustainable recycling the standard in a circular economy.