Packaging giant Tetra Pak is channeling US$1.8 million into advanced artificial intelligence systems at three United Kingdom recycling facilities, a move designed to significantly improve the sorting of used beverage cartons and bolster the nation’s recycling infrastructure. The investment partners the Swedish-based company with Recycleye, a British startup specializing in AI-powered computer vision for the waste industry, aiming to more effectively separate cartons from the complex stream of mixed waste materials.
The initiative addresses a critical bottleneck in the circular economy for packaging: the gap between collection and processing. While nearly 75% of UK councils now collect beverage cartons from the kerbside, the capacity of recycling facilities to accurately sort these multi-material items has lagged. By deploying technology that can identify and separate cartons with precision, the project seeks to increase recycling rates, enhance the quality of recovered materials, and create a more commercially viable system for turning old cartons into new products, aligning with ambitious new national recycling policies.
A Technological Leap in Waste Sorting
The core of the investment is the installation of Recycleye’s QuantiSort system, which combines sophisticated machine learning with high-speed physical sorting machinery. This technology moves beyond the limitations of traditional sorting methods by using cameras to capture images of waste items as they move along a conveyor belt. An artificial intelligence algorithm, trained on a vast database of waste images, analyzes these visuals in real time to identify the unique characteristics of a beverage carton. Once a carton is identified, the system activates a series of pneumatic valves that use targeted jets of air to eject the item from the main waste stream, separating it for its specific recycling process.
How AI Outperforms Traditional Methods
For years, automated sorting has relied heavily on technologies like near-infrared (NIR) spectroscopy, which identifies materials based on how they reflect light. However, NIR systems often struggle to correctly identify composite packaging like beverage cartons. A typical carton is made of approximately 75% paperboard, 20% plastic, and 5% aluminum, and this layered structure presents a confusing spectral signature that does not clearly represent any single material to a NIR sensor. Furthermore, NIR technology is famously challenged by dark or black plastics, which absorb light and return little to no signal for the sensor to analyze.
In contrast, AI-powered computer vision operates much like the human eye, but at a superhuman speed and consistency. The Recycleye system is touted as being as accurate as a person, capable of making up to 1,000 ejections per minute. It can perform granular identifications that are impossible for older technologies, such as distinguishing between an aluminum beverage can and an aluminum aerosol container, or identifying a carton even if it is partially obscured or crushed. This high level of accuracy ensures a purer stream of sorted cartons, which is essential for high-quality recycling.
Unlocking the Value of Composite Materials
The primary challenge and opportunity in carton recycling lies in its composite nature. The initial stage of recycling easily separates the paperboard fibers through a pulping process, and this recovered pulp has a well-established market for making new paper products like cardboard or paper towels. However, the process leaves behind a mix of polyethylene plastic and aluminum foil known as PolyAl. For a carton to be fully recycled and contribute to a true circular economy, this PolyAl must also be recovered and repurposed.
By ensuring more cartons are correctly sorted in the first place, the AI technology provides a greater and cleaner supply of material for the subsequent stages. This enables the viable recycling of the PolyAl, which can be transformed into pellets or boards. These secondary raw materials are surprisingly versatile and are used to manufacture a range of new products, including durable plastic crates, shipping pallets, roofing tiles, and even large-scale 3D-printed items. Through partnerships with recycling specialists like Veolia and Recon Polymers, Tetra Pak is helping to build a stable market for recovered PolyAl, an effort that could double the total economic value of a recycled carton.
Strategic Deployment Across the UK
The US$1.8 million investment is being strategically deployed across three key materials recycling facilities (MRFs) in the United Kingdom. The first Scottish installation of Recycleye’s QuantiSort system is at Levenseat Resource Management in Central Scotland. This is complemented by two other sites in England: J&B Recycling in Hartlepool and Cumbria Waste Management in Carlisle, which was the first UK facility to use the QuantiSort technology.
This initiative represents Tetra Pak’s largest single commitment to UK recycling infrastructure to date and is a key part of a broader financial strategy. The company has allocated US$3.1 million specifically for improving carton sorting systems in the UK, which is funded from a US$44 million annual global program dedicated to developing packaging recycling worldwide. “We are delighted to be able to demonstrate the power that investment in cutting-edge AI technology holds when it comes to improving the UK’s recycling infrastructure,” said Awantika Chadha, Sustainability Manager at Tetra Pak UK.
Alignment with National Recycling Policies
The timing of the investment is particularly significant, as it coincides with major regulatory shifts in UK waste management. The installations are proceeding as both Scotland implements its Circular Economy Act and England prepares for its “Simpler Recycling” program. The Simpler Recycling initiative, set to take effect for most businesses by March 2025 and for all households by March 2026, aims to standardize the materials collected for recycling across all local authorities. This will end the so-called “postcode lottery” and ensure that the same core set of materials—including paper, card, plastic, glass, metal, and food waste—can be recycled everywhere in England, reducing public confusion and boosting participation.
This legislative push is expected to increase the volume of cartons entering the waste stream. According to ACE UK, the trade association for carton manufacturers, nearly three-quarters of British councils already offer kerbside carton collection. As this number grows and collection becomes standardized, the pressure on MRFs to process this material effectively will intensify. Mandy Kelly, CEO of ACE UK, stated that with collection expanding, it is “vital that there is investment in MRF infrastructure.” Tetra Pak’s AI investment is a direct response to this need, creating the processing capacity required to handle the increased supply of collected cartons.
The Broader Vision for a Circular Future
Ultimately, this investment in AI represents more than just a technological upgrade for a few facilities; it is a foundational step toward a fully circular economy for beverage cartons. By solving the complex sorting challenge at the source, the project enables the entire recycling value chain to operate more efficiently and profitably. It ensures that valuable paper, plastic, and aluminum resources are captured and returned to the supply chain instead of being lost to landfills or incineration.
The collaboration between an industry leader like Tetra Pak and an innovator like Recycleye showcases how targeted technology can address stubborn environmental problems. It validates the principle that waste is simply a resource in the wrong place—a core belief of the Recycleye team. As these AI systems become operational, they will not only divert thousands of tonnes of material from disposal but will also serve as a powerful blueprint for how automation and intelligence can accelerate the global transition to a more sustainable, circular model of consumption and production.
