AI and recycling: taking the waste out of waste management

Every year, the world generates over 2 billion tonnes of solid waste. That’s more than the cumulative weight of every single car on the planet — considerably more, in fact. However, only 20% of this is recycled effectively. Given the universal acceptance that recycling has a vital role to play in combatting climate change, that figure shows that something, somewhere, is going wrong.

Global waste production will continue to rise (more people = more stuff being thrown away) — experts predict 3.4 billion tonnes by 2050. So, it’s obvious that we need to get better at sorting and recycling our rubbish if we’re to mitigate the environmental impact of poorly managed waste. So here it comes again… everyone’s favourite two letters, AI.

?Let’s take a look at why more and more companies are turning to automated processes in all aspects of waste management — from collection to sorting to recycling.

?

Only human, after all ?

?The crux of it is that humans sometimes make mistakes. And the technology — in this context, at least — largely doesn’t.

?Sorting the recyclable from the non-recyclable is painstaking and monotonous, especially given the large volumes at play. So, some human error is inherent. Even with the best intentions, people may struggle to correctly identify and segregate various types of materials, resulting in contamination and reduced recycling rates.

?There’s also the issue of inefficiency in collection. Conventional waste management costs a lot, takes a lot of time, and expends a lot of resources. According to Earth.org, unnecessary pickups increase?annual collection expenses by 70%. Route congestion also raises the volume of petrol consumed to carry out the collections, contributing to an?increase in carbon footprint by 50%.

?All of this means more and more municipalities and waste management organisations are turning their gaze towards smart solutions that reduce these inefficiencies and minimise mistakes.

?

The route of the issue

Traditional waste management systems often rely on fixed schedules and routes — as we all know, the pinnacle of domestic peak performance is knowing which bins are being collected and when. But the waste management pros are now realising that these routines can lead to overfilled or underutilised bins. AI-powered solutions address these issues by using smart sensors and data analytics to optimise collection routes and schedules.

For instance, AI-enabled smart bins equipped with sensors can monitor fill levels and notify waste collection teams when service is required. This approach reduces unnecessary pickups, saving time and fuel. Waste management experts Sensoneo claim that data-driven decision making and optimisation?of waste collection routes, frequencies and vehicle loads can reduce waste collection?costs by at least 30%, and carbon emissions up to 60% in cities.

?

Getting sorted

?In addition to collection, sorting and recycling are areas of waste management that are primed for AI-powered improvement, especially given the growing prevalence of robotic sorting systems.

?These systems use machine learning algorithms, computer vision, and sensors to identify and sort materials with impressive speed and precision. For example, AI robots like those developed by AMP (check out their Hollywood trailer-style video) can sort recyclables with 95% accuracy, far surpassing the approximately 70% accuracy of human workers. These robots can distinguish between plastics, metals, paper, and even different grades of materials, ensuring cleaner output streams.

?AI also enables real-time tracking and monitoring of recycling operations. Advanced analytics platforms analyse data from sorting machines to flush out inefficiencies and tighten up processes. Cue significant cost savings and improved resource utilisation. This was proven in practice by Alameda County Industries — a waste management service provider in California — who reported a 59% reduction in labour costs since introducing RecycleOS, a product provided by EverestLabs which is similar to that of AMP. ?

But it’s not just about accuracy and manpower. AI helps tackle contamination too — a major challenge in recycling. The reasons we’re encouraged to painstakingly rinse out chopped tomato tins or yoghurt pots is because food residue can render entire batches of recyclables unusable. AI algorithms trained on large datasets can detect and remove those contaminated items more effectively.

?

Replacement or enhancement?

Whenever we talk about AI and its benefits, we’re acutely aware of the impact this has on humans. The outlook for people who work in waste sorting facilities can vary depending on who you want to believe.

Tegos, a waste software company based in Germany, take the view that there is still an important role for humans despite the compelling labour cost statistics. In a piece published via industry news source Let’s Recycle, they say:

“AI will not replace individuals. But people effectively utilising AI as a tool to replace certain functions or roles will replace humans who don’t. AI is a powerful tool that can augment human capabilities, automate tasks, and provide valuable insights…

…By embracing AI as a tool and leveraging its potential, individuals can enhance their productivity and decision-making, leading to a more collaborative and productive future.”

This very much positions AI as an augmentation to human processes; a helpful tool that can work alongside us. But as more cost reductions and productivity improvements are credited to smart, automated processes, there’s a contradictory opinion that, given the data around, it’ll only be a matter of time before frontline staff will be phased out.

A Wall Street Journal report concludes that in a lot of cases, waste management robots are objectively and provably better at the job than humans. It also goes on to say that “working in waste management can be a dirty job” and that many recycling centres across the US are facing labour shortages. The insinuation is that yes, robots may replace the human workforce, but that’s OK because it’s a job we don’t want to do anyway.

It's a claim we’re not sure holds up to scrutiny. You’ll have to make up your own mind where you stand, but it’s always something to consider when human employability is on the line. Robots don’t have roofs to put over heads or meals to put on the table.

Emerging trends ?

What other technologies are making a big difference in an evolving industry?

??Blockchain technology is a promising innovation that can enhance transparency and traceability in recycling systems, enabling better tracking of waste materials through their lifecycle. This means higher recycling rates and greater trust among stakeholders as more and more emphasis is placed on driving towards a circular economy.

Biotechnology also has potential; enzymes and microbes that can break down complex materials like plastics into reusable components are being greater understood and applied. For instance, researchers are developing "plastic-eating" enzymes that could significantly reduce landfill waste. Let’s hope those enzymes are hungry.

Experts are also discovering new recyclable materials, like eco bricks and plastic roads. Eco bricks repurpose plastic waste to use as robust and versatile building blocks. In many instances, recycled bricks are formed from plastic bottles combined with other recycled plastic. This new material not only has environmental benefits but also elevates what construction engineers can offer due to the plastic’s insulating properties.

Plastic roads add recycled plastic waste to asphalt mix, not only reducing raw materials but maximising the lifespan of roads and pavements too. These durable and comparatively flexible road surfaces are more resistant to cracks and potholes and reduce the wider carbon footprint of highway infrastructure.

?

R(AI)sing the bar

?It’s clear that AI use cases in waste management have the potential to address that troubling statistic we mentioned at the start of the article — anything that can raise the effective recycling rate higher than 20% is an undeniably good thing. And with more innovations either on the horizon or already in use, it looks like more efficient waste management could move the needle in the greater context of climate action.

?

Did you enjoy this piece? Check out our other blogs athttps://www.point6.co.uk/our-thinking/