The rise in investment and use of Artificial Intelligence (AI) in sorting for plastic recycling may be key to improving one of the greatest challenges in recycling, according to a recent report from Andrea Bassetti, senior analyst for plastics recycling at ICIS.
Sorting plastic is a multifaceted matter that makes it difficult to extract high yields and ensure high-quality feedstocks, which are essential to recyclers.
Sorting is a process that is most often done (at least as a first step) manually, relying on workers to sift through material as it rolls down a conveyor belt. Regions such as southern and central America, Asia Pacific, and Africa heavily rely on manual sorting.
This process works for uniformly shaped items, such as bottles and common container types, but leaves smaller and non-uniformly shaped items undetected with the lack of advanced sorting to often end up in the landfill.
Recent investments by companies such as AMP Robotics, Perfect Sorting Consortium (PSC), and GreyParrot have focused on improving the sorting process. In these efforts, AI technology has shown promising results.
This technology can teach robots to detect plastic waste based on a variety of attributes such as colour, shape, material, brands, and more.
According to the US Plastics Pact AMP Robotics case study, the technology specifically increases the sorting rate by 100%, effectively picking twice as many items per minute as manual sorters. Additionally, the picking is more accurate and consistent.
AMP Robotics is leading the movement in the Americas. The US Plastics Pact study details that AMP’s AI platform becomes smarter and more effective as the company deploys more robots.
They can add limitless subcategories of brand-level material to meet market demand and distribute the functionality to identify and sort it across its fleet.
The company can classify more than 100 different categories and characteristics of recyclables across single-stream recycling, e-scrap, and construction and demolition debris.
In fact, they have extended their object recognition run rate to more than 75 billion items annually thanks to its approximately 300 deployments across North America, Europe, and Japan.
In Europe, the PSC has begun to explore the use of AI in waste sorting as a complement to existing technologies as detailed in the study How AI is revolutionizing waste management by Clean Robotics.
The goal of the PSC over the next two years is to develop and AI decision model that would help separate packaging that is not currently properly sorted.
Ensuring the collaboration of recyclers and producers can contribute to the success of the development of the technology and ultimately the widespread adoption across the packaging, if not other industry, sectors.
The capability of AI to learn about packaging to the specificity of a brand’s packaging can serve as an opportunity to capture more specific packaging and directly influence what is recoverable material.
The potential increase in availability of high-quality feedstocks will in turn improve the volume of recycled materials coming through the supply chain, especially as most markets are currently structurally short in supply.
And for food contact applications, in which packaging is prevalent, high-quality feedstocks are essential but even more limited in supply today, especially across the full range of polymers utilised in this sector.
This technology is yet to be applied at a large scale. The nature of AI’s continuous learning shows that the technology should become an important and applicable solution moving forward.
However, applying this solution in a silo will not be enough to resolve sorting. Not only is there a need for significant investments to apply this at scale but product packaging design must also improve in tandem with to ensure that AI is not only becoming better at sorting existing materials and packaging but also new ones.
If you liked this article, you might also enjoy:
The Lidl approach to packaging sustainability
How did Brazil achieve its 100% aluminium can recycling rate – and can it be replicated in the EU?
Experts have their say on the EU’s Packaging and Packaging Waste Directive revisions
A deep dive into the most important packaging sustainability trends and solutions
No comments yet