Sign up for our newsletter
Leadership / Sustainability

How AI and big data are making recycling more efficient

As countries crack down on plastic waste exports, tech innovation is helping to make recycling more effective.

This week, the UK’s much-delayed Environment Bill returns to parliament. The bill includes a ban on the export of plastic waste to Global South countries, but campaigners are calling for that to be expanded after Greenpeace found plastic waste that had been exported to Turkey being dumped and burned, not properly recycled.

In response to the report, Turkey has banned plastic waste imports, following a similar move by China in 2019. These bans mean that countries such as the UK can no longer outsource the challenge of disposing of their plastic waste in an environmentally friendly fashion, and will need to devise new solutions.

Digital technology will undoubtedly play a role. Tech Nation’s new Net Zero initiative, an accelerator programme for start-ups tackling carbon emissions, is actively seeking participants who are focused on recycling. “The waste system that we currently have is fundamentally flawed,” says Sammy Fry, Net Zero lead at Tech Nation. “Ninety-one per cent of plastic isn’t actually recycled globally, which is insane.”

recycling tech
The UK nominally recycles over 40% of its waste but much of that ends up in landfill. (Photo by Nordroden/Shutterstock)

Recycling tech: using AI to sort plastic waste

Already, companies are using artificial intelligence and big data to make recycling more effective by improving the accuracy of waste sorting, boosting the transparency of the global recycling supply chain, and automating waste disposal advice. Tech Monitor met some to assess the state of the art.

White papers from our partners

Sorting recyclables, particularly different kinds of plastic, is a big challenge in the recycling supply chain. A report by the European Commission on plastic strategy shows that although around two-thirds of all plastic packaging is recyclable, in practice just 40% is recycled. This is due to inefficiencies not only in the collection but also in its treatment and sorting.

UK start-up Greyparrot uses AI and machine learning to sort commercial waste more effectively. Designed for use by material recovery facilities (MRFs), where waste material is sorted and shipped to manufacturers, the company says its technology can identify more than 40 waste product categories as they pass through a waste conveyor belt.

According to Greyparrot, one of the MRFs it works with has achieved 94.5% accuracy in assessing the quantity and nature of materials by its technology. Another of its clients in South Korea, ACI Chemicals, is using Greyparrot’s system in their sorting system to detect contaminants in the polyethylene terephthalate (a type of plastic) line.

A digital twin of the world’s waste

Even when waste is sent to be recycled, it often ends up in landfill. In the UK, for example, 46% of all household waste is said to be recycled. But this only indicates how much is sent away to be recycled, not where it ends up. In fact, much of it is shipped to landfill sites in Southeast Asia.

One reason for this is the complexity of the global waste management system, which involves multiple stakeholders working across many continents. This makes keeping track of waste materials through the recycling process almost impossible.

“You’ve got different materials being generated as a waste in different locations that are picked up and moved by different companies, taken to different sites where they might be processed or transformed or sorted, and then moved on to other locations,” says Michael Groves, founder and CEO of Topolytics. “And at some point, something happened to it. It’s a very complex kind of supply chain.”

What we’re trying to do is make the global waste system much more visible.
Michael Groves, Topolytics

To tackle this problem, Topolytics is integrating various data sources to create a digital “waste map of the world” that can trace the full waste journey, from where it is created to where it ends up. This map offers business and local authorities visibility into what happens to their recyclable materials. “What we’re trying to do is make the global waste system much more visible,” explains Groves.

Groves and his team have created a platform called WasteMap, which constructs a digital twin of waste materials’ movements around the world using data from government and commercial sources. By aggregating multiple sources, Topolytics aims to build trust in the data. “[If] you can build trust in the data… then ultimately you can unlock value in the material.”

Although most of Topolytics’ clients are in the private sector, the company is currently working with the UK government on a prototype WasteMap to help achieve tracking of all waste digitally across the country. Currently, such tracking relies on paper-based systems.

“If you think about this, across the world more than 60% of waste material is still going into landfill or a waste dump or it’s leaking into the environment,” says Groves. “We think there’s a big opportunity to use data at a big scale, a sort of global scale, to really improve the environmental and the commercial outcomes for that material.”

Reducing the creation of waste

Another company using technology to make waste management more environmentally friendly is US-based Recycle Track Systems (RTS). The company has produced an app that advises businesses and individuals on the best way to dispose of materials. The app uses tech to analyse pictures of a given material and advises the user what they can do with it, whether it is donating it for reuse or sending it to particular recycling facilities.

“Our goal is to make sure that materials do not make their way into landfills and therefore reducing our impact from climate change,” says Allyn Shaw, president and CTO at RTS.

Bans on plastic waste exports present the US with an opportunity to transform its waste infrastructure, says Shaw. “Given that China and other countries recently banned the exportation and importation of recyclables from the US, we need domestic solutions based on technology,” says Shaw.

However, the current US infrastructure is far from efficient. “These facilities need sensor technology and optical sorting: innovative technology that brings us up into the 21st century, and that’s going to require serious investment.”

We talk a lot about recycling, but that’s a reactive point. What can we do to reduce the generation of waste to begin with?
Allyn Shaw, RTS

But while tech making recycling more efficient is essential, Shaw argues that more investment and innovation should be directed towards reducing the creation of waste in the first place. “We talk a lot about recycling, but that’s a reactive point,” Shaw tells Tech Monitor. “What can we do to reduce the generation of waste to begin with?”

That is why instead of “garbage”, RTS prefers to talk about materials and material management, Shaw says. By recirculating materials through reuse and recycling, the overall generation of waste is reduced.

Tech Nation’s Fry agrees that greater focus should be given to reusing the materials we consume, to reduce the environmental burden from waste. “I think reuse at the moment is a really hot topic and something that’s really worth focusing on, because if you get it right, it’s far more efficient than recycling because rather than having to completely repurpose and redesign material, you are just reusing the material that you have. And I think there’s a lot that the government could do there as well.”

Cristina Lago

Associate editor

Cristina Lago is associate editor of Tech Monitor.