Matanya Horowitz is the founder and CEO of AMP Robotics, a Colorado-based company in the recycling infrastructure space working to apply AI and robotics for commodities recovery.
The last year brought no shortage of attention-grabbing headlines about the ways recycling is flawed, or how we’ve been misled into believing it works. To be clear, recycling was never intended to be the sole solution to the proliferation of waste. But what these commentaries fail to capture is the headway that technology-driven efforts are making in shifting industry economics and increasing recycling rates.
Investment in recycling from venture capital and private equity firms, coalitions, government entities, and consumer packaged goods companies validates the role advanced technology is playing in addressing the industry’s challenges — and the broad conviction that exists in elevating business models focused on more sustainable materials management.
As 2023 gets underway, AI will lead trends in waste technology and continue its role in advancing the strategic objectives underlying the U.S. EPA’s road map to achieving a 50% recycling rate by 2030.
AI’s precision expands markets for recycled commodities
Technology’s undisputed role in unlocking material markets is poised to continue. AI’s precision allows for the separation of many categories of materials that opens doors for the industry.
Take PET thermoforms, a material typically made from No. 1 plastic that is often used in clamshells and berry containers. There are often limits on the amount of PET thermoforms that can be commingled with bales of more valuable PET bottles. On top of that, low prices for virgin resin create a competitive challenge. And while buyers of PET bottles are willing to pay a premium, that’s not always the case with other PET markets like thermoforms.
But AI technology like ours can separate thermoforms from other PET materials for both mechanical and advanced recycling, enhancing their value. The technology does this with a simple software update to the existing install base of robotic systems. These improvements in technology capabilities are creating end markets and improving circularity for this historically challenging material.
AI, applied in primary or secondary sorting environments, allows recycling operations to create very specific product blends for different end markets, and to update these blends as demand and markets evolve.
AI strengthens and reinvents recycling infrastructure
AI will continue to enable new types of sorting devices for the recycling industry — driving infrastructure improvements as it supports end market development. Thin film materials like grocery bags and stand-up pouches, for instance, have long been an issue for MRFs around the world. They significantly impact the maintenance of screens and the quality of recovered material — fiber, in particular. It’s rare I see a MRF that doesn’t have a challenge or headache dealing with film somewhere in the system; more typically, it’s multiple places in the system.
AI is laying the groundwork not only to reduce the contamination burden on MRFs, but to scale the recycling of film and flexible packaging. Many MRFs are looking for a solution for film contamination immediately, even if it’s simply removing the film to residue for the benefit of keeping it out of screens and fiber bales. With the ability to capture film, even if it goes to residue initially, demand will be easier to catalyze.
Flexible packaging has been almost uniformly single-use, but major brands continue to make commitments to use more recycled content in their products, and several states have recently adopted laws aimed at ramping up the use of postconsumer resin in plastic products and packaging. AI is addressing how to recover film economically, which will boost recovery and demand for products manufactured from recycled film and flexibles to develop and support end markets.
AI redefines contamination in the materials stream
AI will continue to reshape what’s considered contamination. AI continuously learns to recognize new types of packaging as brands and manufacturers introduce them into the recycling stream and deploys the learnings to the entire installed base. The vast amount of material being identified, categorized and handled demonstrates that most of the materials found in a curbside stream can be sorted and processed.
The ability to train AI on new packaging types and formats and transmit that information fleetwide will allow the industry to keep up with the breakneck pace of innovation packaging producers and manufacturers are releasing to store shelves. Technology is flipping the script on materials long considered contaminants, like thin film. If an item can be sorted, then recyclers have a broader array of options for what to do with that item. The notion of “hard-to-handle” materials will be relegated to the waste bin.
AI centralizes, proves value in data
AI will help the industry overcome inconsistent access to and use of recycling data. With AI and computer vision, we’re able to identify and describe the objects that pass through MRFs and transform them, both desired commodities and undesired contaminants, into data. Digitizing the waste stream provides transparency about what is and isn’t getting recycled to create better solutions on the path toward a circular economy.
AI technology can record the number of items of a desired commodity type and derive additional data from those objects. This, combined with the mass of produced scrap bales, can offer detailed insights into the trends of a specific material flow — say PET bottles or PET thermoforms — in a given MRF, a network of MRFs, or even across a geography. This is powerful data for planning and understanding the supply of desired materials.
At the facility level, the data AI generates can help recyclers scrutinize their operations, identify inefficiencies and take instantaneous action to rectify them. Material characterization solutions provide actionable, data-driven insights throughout the facility and provide the tools necessary to centralize data collection and reporting.
AI enhances policies to support recycling
AI systems will also revolutionize the way policy makers view the recycling industry. Over the past two years, four states — Maine, Oregon, Colorado and California — have passed laws related to extended producer responsibility for packaging, and many other states are considering EPR-related bills.
As these programs continue to ramp up across the country, the producer responsibility organizations (PRO) as well as the regulators tasked with overseeing these programs will be faced with an immense challenge. That is, how to collect, process, and make sense of the vast array of packaging types and brands for which manufacturers are responsible. Because AI systems already broadly deployed are identifying and categorizing the material coming down the line, it’s only logical to tap this data repository for information the PRO needs to manage EPR programs required by state legislation and regulations.
Certainly, there’s a big difference between what’s recyclable and what gets recycled. We as a society certainly can — and must — recycle more. In addition to commodity prices, factors like markets for recovered materials, access to recycling programs, reclamation capacity and more all affect recovery rates.
Again, recycling was designed to work in concert with other waste management strategies like reducing consumption and reusing material. But claims that recycling "doesn't work" or "can't ever work" fail to account for the high degree of innovation making recycling processes more efficient and cost-effective in 2023 and the years to come.
Contributed pieces do not reflect an editorial position by Waste Dive.
Do you have an opinion on this issue, or other topics we cover? Submit an op-ed.