TOMRA and NEXTLOOPP are currently working on a project that uses cutting-edge artificial intelligence technology to make critical recycling processes more efficient. In this article, Professor Edward Kosior founder of Nextek and NEXTLOOPP, tells us more about this groundbreaking collaboration.
It is almost four years since we launched our ground-breaking multi-participant project, NEXTLOOPP, to close the loop on post-consumer food-grade Polypropylene (PP) packaging.
The world was still reeling from the emergence of COVID and travel, let alone human interaction, was limited. But Nextek had already spent over a decade researching and trialing science-based solutions to accelerate efficient sorting and powerful decontamination of what we had identified as a key missing link in the recycling stream, food-grade PP.
We were ready to launch our cutting-edge technology in the commercial world. The timing was perfect, in 2022 the global PP market was calculated at £60bn, projected to rise to £70bn by 2028. Yet only 1% of this is being recycled.
In Europe alone, 44 percent of PP is used in consumer packaging, split between rigids (22 percent) and flexibles (16 percent). Brand Owners were looking for low-carbon solutions to the circular economy.
The sorting dilemma
NEXTLOOPP’s progress since 2020 has been well documented. The project’s innovative fluorescent marker technology was one of the main contenders to address the issue of differentiating food-grade plastic packaging from the waste stream.
However, we were not alone. There were a number of cutting-edge solutions aiming to achieve the same thing resulting in a lack of consensus as to which technology should be adopted.
The common drawback was that marker technology, by its very nature, requires changes to the labels or the packaging, and industry as a whole was at odds regarding which technology should become the standard.
NEXTLOOPP’s 50 global participants have been trialling and deploying Nextek’s technologies for sorting as well as decontaminating, keeping an open mind as to which technology would best serve the industry.
A clear belief that authentic collaboration can accelerate results recently paid off when the project achieved a major sorting breakthrough that has resolved the dilemma of which sorting technology should be universally adopted.
Beyond expectations
Since the start of NEXTLOOPP, one of the main focuses was to efficiently separate food from non-food packaging for which the team very successfully trialled UV markers in conjunction with NEXTLOOPP participant, TOMRA.
At the time this was the most effective spectroscopic sorting technology to separate the same polymer into food and non-food fractions by adding an additional, fully integrated, coded sorting dimension to the standard NIR/VIS sorting systems.
Even prior to NEXTLOOPP launching, sensor-based sorting technology leader, TOMRA had introduced the industry’s first AI-based deep-learning sorting solution to separate silicone cartridges from PE streams and later for wood sorting (in 2019).
In early 2024 TOMRA introduced the food-grade application to address the industry-wide challenge of food-grade separation.
NEXTLOOPP supported the PP field validations conducted by TOMRA to test GAINnext’s capabilities in industrial conditions, however given that the AI had to be taught and NEXTLOOPP already had our own highly efficient marker technology plug-and-play ready, we believed we would start with markers and then phase in AI.
We did not anticipate the incremental speed at which TOMRA grew GAINnext’s capabilities.
By early, 2024, TOMRA had accelerated its GAINnext deep learning technology to separate food and non-food plastics and were using it to identify PP packaging, amongst others. It did not take long to realise this was a real game-changer.
The system correctly identified over 95% prior food packaging content, an outstanding result that is poised to enable brands to meet the sorting standards required to deliver the food safety authorities’ stringent requirements.
AI-boosted sorting
Since those initial trials, TOMRA and NEXTLOOPP have run a series of ground-breaking trials using TOMRA’s near-infrared, visual spectrometry system AUTOSORT combined with their latest deep-learning technology GAINnext to show how deep learning, a subset of AI, could supersede markers for rigid packaging and fully resolve the food-grade PP sorting hurdle.
During the latest full-scale trials, AUTOSORT with GAINnext sorted five tons per hour of mixed PP plastic packaging and exceeded 97 percent food-grade content in the sorted output.
This development is an invaluable boost to NEXTLOOPP whose participants confirm that TOMRA’s new sorting system has the potential to be rolled out to all PP packaging sorting facilities since it focuses on the packaging design attributes rather than any form of additional markers.
Accelerating recycling’s next step - decontamination
By providing a sorted food-grade PP PCR stream, AUTOSORT with GAINnext can now accelerate the supply of food-grade rPP via the NEXTLOOPP decontamination process in many more recycling operations globally without any further delays associated with new label or marker requirements.
As a consequence, this breakthrough will positively impact production of valuable food-grade PP PCR streams.
Transitioning from markers to AI
Less than 12 months ago NEXTLOOPP’s focus was on packaging design guidelines to facilitate sorting packaging into single-polymer fractions using markers.
Now TOMRA’s latest innovation has flipped this element of the design guidelines on its head. Instead of a system that relies on labels featuring specific markers, the neural network of the AI system is trained to identify a wide range of shapes and packaging attributes. Through structured training, it learns to separate out food contact from non-food-contact packaging.
Design for deep learning
The next step is to revise current design guidelines to take into account how the AI ‘thinks’ to continuously enhance both GAINnext’s and other existing sorting solutions’ capacity. Certainly the suggested changes to the packaging will be simpler and more cost-effective than relying on labels and markers, if anything the more stereotypical the packaging, the better.
The principles by which GAINnext recognises a package are based on object recognition. By segmenting a range of different design factors of the pack the AI gathers the different triggers to build its contextual memory of every pack it is shown.
Using the road sign analogy, whereby the iconic stop sign is internationally recognised, the AI is trained on food package shapes, sizes, dimensions or other criteria that frequently re-occur. Transparency, opacity, print, shapes and colours alert the system that is designed to aim for accurate recognition of the sorted PP packaging.
Next gen packaging
The more stereotypical the pack shape, inclusive of readily identified attributes, the higher the rate of identification. Given that PP food trays are predominantly unpigmented or white and rectangular, these are easily picked out.
The likes of ice cream tubs, however, which often are solid white, are likely to be rejected from the food packaging stream as they could be identified as non-food dishwasher capsule packs. This is where design attributes and deeper learning can boost the correct recovery.
This brings us back to NEXTLOOPP’s original suggestion of using colour or design features to signal whether a pack belongs in the food or non-food category. Using colour or design features to identify a pack’s former use would enhance AI capacity to define the pack’s destination.
By making packaging as easy to identify as possible, brand owners now have the opportunity to signal their pack’s recycling identity without relying on a label. This has huge implications in data collection during sorting and opens up the opportunity for Extended Producer Responsibility reporting and recovery of discrete packaging streams for brand owners via novel PRF configurations.
GAINnext’s training is such that even in a scenario where the pack is crushed, torn or otherwise damaged, it can pick up enough attribute points of differentiation to make an effective sorting decision in a split second.
This AI system is rapidly building a wide range of cues to identify and differentiate packaging and this is an ideal opportunity for brand owners to adjust their packaging to align with the way that AI is ‘thinking’.
This breakthrough in high-speed and precise AI sorting, exemplified by TOMRA’s AUTOSORT featuring GAINnext Deep Learning technology, is already making a significant impact on the recycling industry. As traditional markers become obsolete for rigid packaging, sustainability and circularity will soon be effortlessly incorporated into packaging design, delivering both environmental and economic advantages.
If you liked this story, you might also enjoy:
The ultimate guide to the Packaging and Packaging Waste Regulation in 2024
How are the top brands progressing on packaging sustainability?
Sustainable Innovation Report 2024: Current trends and future priorities
Everything you need to know about global plastic sustainability regulation
No comments yet