The Rise of Recycling AI

Prairie Robotics, a Canadian firm that offers artificial intelligence-enabled camera systems for recycling collection trucks, is expanding its partnerships with municipalities in North America that want to reduce contamination in their curbside streams. The company works with about 40 cities in the U.S. and Canada.

The company recently partnered with the City of Tacoma, Washington, on a two-year project to help educate residents on what can and cannot be recycled in the city’s curbside program. It’s funded by a $1.8 million grant from the U.S. EPA, announced in 2023 as part of the agency’s Recycling Education and Outreach grant program.

Prairie has also recently launched partnerships in Centerville, Ohio; Livermore, California; and Springdale, Arkansas, “with more that are in the pipeline,” said CEO Sam Dietrich. The company also works with haulers such as GFL Environmental and Waste Connections, who use the tech’s data to track metrics including bin set-outs or overfilled containers.

Collection vehicles have long had cameras meant for safety and fleet management reasons, but Prairie is among the companies capitalizing on a newer use for AI-enabled on-truck cameras: measuring recycling contamination in real time. 

The system can detect contamination, such as food waste or trash, at the moment the material enters the recycling truck. Then, it determines the origin of that contamination and can send customized messages or postcards to residents at the associated address with pointers on how to recycle correctly. 

The technology has gathered momentum in the last few years as municipalities and private haulers both grapple with contamination-related costs and seek better ways to achieve cleaner, higher-value recycled commodities, Dietrich said.

Through its use of Prairie’s cameras, Tacoma aims to gain insight into residents’ recycling habits and use the data to better tailor recycling and contamination messaging

“Contamination impacts how we can deliver services and the cost of those services for all residents,” said Lewis Griffith, Tacoma’s Solid Waste Management division manager, in a news release announcing the program. “Educating residents on what is accepted in our curbside program, and reducing contamination, can have positive impacts on the program for all residents of Tacoma.”

The pilot will roll out to residential recycling routes in phases over the next year. Tacoma officials say some neighborhoods will begin receiving postcards starting this month, but residents won’t face fines or other penalties for contamination during the pilot. 

The project will continue “through the life of the grant” and is expected to be reviewed in June 2027, according to the city’s Environmental Services Department.

Dietrich said Prairie Robotics can tailor its “toolbox” of services for each municipality’s needs. Tacoma wanted the ability for its route supervisors to manage and monitor contamination within the areas they’re assigned, as opposed to some other cities that examine the data on a citywide or regional basis, he said.

“Every municipality is running a very similar program in terms of what the deliverables are, but they're all running it in their own unique way and use case,” Dietrich said. 

Some municipalities mainly want Prairie’s services as a way to send personalized postcards to the households with the highest contamination rates, while other cities have added a feature that sends a formal warning letter after three strikes. “It gives municipalities the ability to, through our platform, automatically escalate that education,” he said.

Some municipalities are more interested in tracking specific kinds of recycling, organics or green waste diversion, or tracking such diversion rates during set periods of time, he said. In Olympia, Washington, the city accesses that data from a mobile app, which helps determine where employees will go to conduct in-person “lid lift” audits. 

These audits are for households “with those repeat, continuous issues that don’t improve with a postcard alone,” he said. “I think a lot of people see our technology is fully automated... But for communities that explicitly want to continue that feedback [in person], we're happy to support that, too.”

Haulers with commercial contracts have also used the technology to help narrow down where specific contamination might originate from on a building to building basis, he said.

“What we've seen is that waste haulers want to see this data in real time. They want to be able to have their teams engage commercial clients the same day to resolve these issues,” he said.

Many of the projects start as a pilot. Dietrich said the majority of the cities where Prairie operates are now on a long-term, recurring contract. It still operates in Regina, Saskatchewan, the location of Prairie’s first project.

Prairie is also expanding its technology to be more integrated with other recycling technologies and apps, he said. The company can now sync its technology with apps like Recycle Coach, which helps residents set up collection day reminders and offers sortation pointers for a particular region. 

In the Okanagan region of British Columbia, for example, Recycle Coach subscribers will receive in-app notifications from Prairie based on contamination data from their address.

Policy shifts and brand commitments are boosting demand for food-grade polypropylene that can be used as recycled content in packaging, but the supply is limited, says a report by Closed Loop Partners. Using artificial intelligence-enabled equipment at MRFs aids material recovery at scale, according to the report, which covers results from a study done in collaboration with technology company Greyparrot.

Clear and white food-grade PP, such as beverage cups and yogurt tubs, is abundant in recycling streams, representing more than 75% of the PP captured at the four MRFs in the study, according to the report. 

AI-enabled technologies provide characterization data for huge amounts of material at a more granular level than what was previously available, which contributes to increased recovery and value creation, the report says.

This study, led by CLP’s Center for the Circular Economy and the Closed Loop Foundation, is described as one of the most extensive material characterization initiatives so far. Understanding what’s in a PP bale is the first step toward meeting demand for the material, according to the report, which is why the partners embarked on the characterization study.

They analyzed more than 650 tons of materials at four U.S. MRFs that already accepted PP material. The team installed Greyparrot’s AI-enabled analyzer technology at each location. Participating MRF operators included Balcones Recycling in Texas, Cougles Recycling in Pennsylvania, Rumpke Waste & Recycling in Ohio and Eureka Recycling in Minnesota

The analyzer units were installed above moving belts at MRFs. They recorded an image of each object that passed along the belt and generated classifications that included object mass and area, as well as the likelihood of the item being food-grade material. Greyparrot’s system included classifications for 20 PP items, including white and black containers and lids.

AI equipment and more granular data collection increases transparency, the report says. Plus, it offers opportunities for MRF operators to experiment with creating different bales, such as bales dedicated to food-grade PP. That can boost quality and create more value, the report says.

White PP was about a third of the material examined at the MRFs in this study, and clear PP was about half. Beverage cups were the most prevalent clear packaging, at 31%, followed by lids at 27% and pots and tubs at 20%. While nearly all the clear and white PP in the study was food-grade material, about 90% of colored PP was likely made of non-food-grade material.

The study specifically worked to uncover data at one point in the PP recycling chain, and the next step is to leverage such data to create downstream opportunities, the report says. This could help to identify and break up bottlenecks that prevent food-grade materials from reaching their highest and best-use end markets. CLP said this study lays the groundwork for additional material recovery studies, and future work could focus on other materials besides food-grade PP.

“We look forward to identifying more opportunities to pull valuable food-grade materials back into food service packaging supply chains –– a critical step toward recycled content goals and packaging circularity,” Kate Daly, managing partner and head of the Center for the Circular Economy, said in a news release.

The study was co-funded by the CLP-managed NextGen Consortium, which includes major companies such as McDonald’s, Starbucks, the Coca Cola Co. and PepsiCo.

Artificial intelligence is changing the face of waste education. City staff and facility operators are looking beyond just paper mailers and public service announcements to cameras, screens and text prompts. The technology is designed to meet users at the bins and help them navigate the complexities of local recycling rules. 

AI has been used to identify items and trigger physical separation within MRFs for years, and more recently it’s even been implemented to track contamination levels at the point of collection when haulers integrate camera systems into their trucks. But neither of these use cases go as far upstream as trying to help consumers make the right decision about where to put their refuse, recyclables or organics.

“This was really an effort to help people better understand small nuances, and it also gave us new insights about common questions and gaps in knowledge,” said Pamela Perez, the marketing manager for LA Sanitation & Environment, which recently introduced Professor Green, an AI tool for organics recycling education. 

Proponents of these AI feedback systems hope they can cut down on contamination and increase capture rates, but the technology is still evolving.

Whether the tools are chatbots or item-identifying cameras, operators have to tinker with the software over time and hone its ability to offer users the right responses. Experts say that accurate item identification — particularly at the curb, where trash and recycling come in an almost infinite number of forms — could be a challenge for some iterations of this consumer-facing AI. Another hurdle will be convincing consumers to use it.

Send a text or take a photo

Contamination rates remain stubbornly high, as recycling rates are also stagnant in many parts of the U.S.

In New York City, for example, non-recyclable contaminants in the metal, plastic and glass waste stream have gone from nearly 34 pounds to almost 49 pounds per household between 2017 and 2023. The Recycling Partnership estimated in 2020 that the inbound material contamination rate was 17%. 

Waste and recycling professionals hoping to improve these statistics are looking to AI as enabling coaching for residents and visitors on the spot. 

Some public-facing AI sorting systems identify items the same way MRF technologies do. Oscar Sort, a platform offered by Intuitive AI, uses a camera to see what a user is holding. A follow-up message on a mounted screen shows which bin to drop it in — trash, recycling, compost, or even (as one client requested) a chopstick-only receptacle for a recycling service to turn into wooden furniture. The software can also offer multi-part instructions, like directing a coffee cup to the trash bin and a cardboard sleeve to the recycling bin.

In the U.S., facilities such as the San Francisco Ferry Building and Seattle-Tacoma International Airport are in the early stages of using Oscar on site. The company says over 25 airports and 50 universities asked to become customers in the last year.

Each Oscar installation costs $10,000 to $15,000, after which yearly services and data fees cost $5,500 to $6,500, said Brian Sano, head of strategy at Intuitive AI. In 2024, the company launched an ad-based offering — brands can pay to advertise on the embedded screens, generating potentially enough revenue for the facility to cover annual maintenance costs and then some.

Consumer-facing sorting AI can also take in what the public offers via text, which is how the city of Los Angeles opted for its technology to operate.

Professor Green, the AI educator built for the city by software company Hello Lamp Post, works like a chatbot. Users in one of four pilot neighborhoods — Porter Ranch, Northridge, South LA and Watts — scan a QR code or go to the city website to text Professor Green. The system replies to tell users if their item can go in a city-provided green bin, or their trash or recycling bins instead. The chatbot understands and replies to English, Spanish, Korean, Tagalog, Ethiopian and Armenian — languages selected to best serve populations in the pilot area.

Ideally, Professor Green clears up local confusion about composting rules. The City of Los Angeles has a lot of residents who move in to the municipality and might not be familiar with local guidelines, Perez said, or need help parsing the rules from those of nearby areas. For example, some of the other municipalities in Los Angeles County asks that residents line their bins with compostable bags, but the city of Los Angeles only accepts paper-product liners.

Too few trash photos

Both Intuitive AI and LA Sanitation staff dip into the data to make sure that their programs are identifying items correctly.

Intuitive AI does this for Oscar customers by comparing the instructions Oscar offered to the recordings of what the user was holding, every day, along with four waste audits a year for most customers.

LA Sanitation staff also comb through Professor Green requests every day. They spend at most an hour on the inquiries, which is often more than enough time to spot any issues. For example, staff taught the AI to respond to questions about tamale waste, but soon realized it didn’t know corn husks were part of the food. Perez said the team has since instructed Professor Green to associate the two.

Text-based sorting inquiries will likely give users more accurate instructions than camera-based technology, said Sharon Hsiao, an assistant professor of computer science and engineering at Santa Clara University. 

That’s because having people tell the AI what they’re trying to dispose of gets around the issue of having to train the software to recognize what trash looks like. Written prompts can still run a wide range. As Perez pointed out, her office had a long list of terms for “pet waste” that Professor Green needed to recognize. But visual identification has even more variability. 

“Cigarette butts or COVID masks — those are very distinct,” Hsiao said. “But if you crumple up paper, it won’t always be the same shape.” 

There are also few photos of trash that are freely available for AI models to learn from.

Hsiao and her colleagues tried to compile a database of these publicly accessible training images when building an AI waste sorter that relies on a smartphone camera. Not only were there too few images to train the software, but the computational power that would be needed for the AI to recognize any possible item goes beyond anything a smart phone can handle, Hsiao said.

An AI system focused on waste requires a small computer and is more likely to be accurate if developers can train the software to recognize a limited array of items, Hsiao said. Places that restrict the kinds of materials allowed inside, like entertainment venues or airports, could be a natural fit. 

Lower barriers are better

Experts expect technology to play an increasing role as local governments and facility operators look for ways to evolve their recycling education. But figuring out what will resonate with the public is an ongoing process.

AI-enabled camera systems might be easier for users, despite being harder to teach. In surveys, Hsiao and her team found that people are likely to think of texting or responding to voice prompts as more work than waiting for a camera to register what it’s seeing. And all of these options were considered significantly more time-consuming than the status quo —the split-second decision of throwing items into whichever bin it seems to belong.

“All the people that we surveyed wanted to be able to use their phone to scan items to tell them which bin it's supposed to be in,” Hsiao said. “In reality, they actually don't do it.” 

To convince people to take the time and use the AI identification they claim to want, Hsiao and her colleagues are working on a sorting system that also tells users about the benefits of their actions. Her team is experimenting with feedback about money or greenhouse gas emissions saved via correct sorting.

For now, the simpler text-based Professor Green option appears to be resonating. Los Angelenos have had over 4,000 conversations with the AI since its debut in March. Use has been particularly high in South LA, a low-income neighborhood where the majority of households speak Spanish at home.

LA Sanitation is now looking at expanding Professor Green to other neighborhoods in the future. But to know for certain whether the AI system is meaningfully changing contamination rates or diverting more material from landfills, Perez and her team would have to see the results of a more old-fashioned tool: A waste audit. 

This story first appeared in the Waste Dive: Recycling newsletter. Sign up for the weekly emails here.
 

WM President and Chief Operating Officer John Morris remembers driving his first waste vehicles in college, a time when riding on the back of the truck was standard and “the only air conditioning was rolling the windows down.”

Today, Morris oversees field operations for the company, which has a fleet of over 18,000 collection vehicles and a suite of technologies to make those vehicles safer and more efficient. “The technology that's in these trucks is light years ahead of where it was,” he said. 

Morris was appointed president of WM on May 14. He’ll continue his field operations role, and multiple executives, including those leading sustainability, customer experience and enterprise strategy, will now report to him. Morris has previously served as market area general manager of New York City, area vice president of the greater Mid-Atlantic Area, chief strategy officer, and senior vice president of field operations. 

Morris says the company has been strategic about how it adopts and rolls out the latest tech upgrades. The industry struggles to retain labor, he said, and the right investments are helping WM become less labor dependent. At the same time, it also provides opportunities to retain employees by providing vehicles and facilities that are safer and more comfortable than before. 

“When we buy more expensive, sophisticated trucks, it costs more to maintain them, but the overall benefit to the business is something we’re comfortable with,” he said. “First and foremost, I want to be safe. But we're also in business to be profitable, not just efficient.”

Morris sat down with Waste Dive on May 6, during WasteExpo in Las Vegas, to discuss how WM is adding AI upgrades to its existing technology, improving on longtime cab technology and piloting new programs for heavy equipment operators. He also discussed pilot programs for using autonomous vehicle technology at landfills. WM CEO Jim Fish has previously called for a “significant percentage” of WM’s heavy equipment fleet be autonomous by 2030.

This conversation has been edited for length and clarity.

WASTE DIVE: AI has become an important element of equipment and fleet improvements in the industry, and it’s been a big topic of conversation at WasteExpo. What’s new on that front for WM?

JOHN MORRIS: AI is helping inform the business better and helping us operate more efficiently. We’ve had forms of artificial intelligence in the business for probably eight or 10 years.

If you go into our recycling facilities, what was once a manual sorting role has been automated over the years. When you look at those optical sorters, that is a form of artificial intelligence. It's reading the material and sorting without us ever having to touch it. AI is also helping us from a safety perspective to avoid things like repetitive motion injuries, as well as from an efficiency standpoint.

When you look inside some of our cabs, it’s interesting to see why [we first adopted AI-assisted technology in trucks]. We had a very prohibitive labor agreement on the West Coast [about 15 years ago], and we were having a problem with our drivers taking pictures if the cans were overloaded. They were supposed to take a picture and send it to a mailbox so we could send the sales rep out [to talk to the customer.] We could not get folks in that particular labor union under that labor agreement to take pictures.

We started looking for ways to augment what they wouldn't do by using technology. What that has morphed into is now every one of our commercial front-load trucks gets still or video images of every transaction.

When we service a container, we can see what's happening as we're going into the container, both the outside and what's inside, and we have cameras on the hopper. We probably pick up about 700 million yards a year of commercial material. That gets us data on every one of those transactions, so you really start to understand customer behavior to see what’s working, not working, or you can address safety issues. 

Virtually none of those images are processed by a person. There's still a human in the loop, but it’s more of a [quality control] thing. We have a team in India, and if we have to send [an image] to somebody in operations or sales, that information will come back overnight, so there's a lot of power in that. 

The other area where I've been thrilled with the progress we're making is really from a safety perspective. We were an early adopter of Lytx [a fleet management and telematics technology]. We have that technology in every one of our collection vehicles and all our supervisory vehicles.

We've been using artificial intelligence and recognition technology to be able to identify [potentially unsafe driving behavior]. That information, gathered through AI, is coaching people in a positive way, as opposed to disciplining them.

It's so hard to hire drivers. So when we look at that technology, we see a coaching opportunity and a development opportunity. You can sit down with one of your drivers and say, “Hey, listen, here's what we've been noticing about your behavior. You should be following at three seconds, because if you don’t, you're 40% more likely to have a collision.”

Like a lot of other haulers in the industry, WM has also been working on its route optimization process. What’s the priority there?

We've had routing engineers and different forms of routing software. Everybody in the industry has for a lot of years, but there was really never a software program that was built specifically for the waste industry. With the number of transactions we do a year, certainly, we want to make sure that our folks are efficient. We also found that when you have “routine breakers,” meaning you have to deviate from a schedule for some reason, the risk of having an incident goes up. 

We run about 4,000 industrial roll-off routes. How we route on Monday is entirely different on Tuesday. So there's a dynamic nature in that industrial line of business that's unique — more so than commercial, and certainly more so than residential. We saw the opportunity to build out an [industrial] routing technology through our internal ops research team. The key is that it's dynamic. There are about 72 attributes that go into making a decision on how to service an industrial roll-off container, things like container type, truck type, disposal type, location. 

We just made another improvement that we're rolling out. We've had tablets in our trucks and onboard computers for over a decade, but when you pull up to a container, drivers have to [interact with the screen multiple times]. What we're trying to do is find ways to make the technology passive. … We want them to focus on safely servicing their customers and not get distracted. So we went back with our software engineers and reworked it so that now we'll use the latitude and longitude of the truck and actuators on the arms to be able to automatically confirm that we've serviced the container.

What’s the overarching strategy for WM when it comes to budgeting for and integrating these kinds of tech and AI improvements? Where do the labor considerations come into play?

I often talk about total cost of operation. [For example,] we have more sophisticated trucks now, with a lot more technology, so they're going to be more expensive to maintain. I'm more interested in knowing the total cost of operation. 

We can look at how we allocate dollars on capital, and what we've been cautious about is making sure that we're not putting technology in for technology's sake. When you talk about AI, it means 50 different things to 50 different people. I talk about it in terms of, how are we going to use technology broadly to modernize our business model and structurally get our cost down? 

Even though the pressure on the labor market has eased from where it was a few years ago, I think over the long term, if you sort of follow the math, those labor scarcity issues are going to persist. I sit on the board of the National Association of Manufacturers, and there are stats [from that business sector] that say in the next five years we're going to need 4 million folks to fill manufacturing jobs. We'll be lucky to get half of them. 

Do you think autonomous vehicles will be part of the equation soon?

There’s something we are piloting right now [in Arizona and in the Pacific Northwest]. We've done some work with some heavy equipment folks on remote control operations.

If you have a facility in the middle of nowhere, a very hard-to-access place, it's gonna be hard to get labor. However, maybe you can put someone in a chair in what looks like a gaming room, they can run the equipment from anywhere. That certainly is going to help from a labor perspective. An example could be automating operations like [loading dirt at a landfill]. 

Heavy equipment is probably a near-term application for those reasons. I could see an application where we might be able to do some form of autonomous movement of long-haul waste. But are we right around the corner from having a fully autonomous refuse vehicle running through neighborhoods and school zones? I think that [will take longer]. 

As far as labor training, we've had training centers for years, and we train about 85 to 90 drivers every week and about 15 to 20 technicians every week. We are just now rolling out the [program] for heavy equipment operators. Our training facility in Florida is on a closed landfill, so it's an ideal spot for heavy equipment operators. 

We are going to continue having labor scarcity and facing the cost of labor, and we continue looking to the advancement of technology [to help] outstrip that labor pressure. At the same time, everybody in the industry is trying to figure out how to hang on to their employees.

Amp will partner with Waste Connections to build and operate a new MRF in Commerce City, Colorado. The artificial intelligence-powered facility, which will be owned by Waste Connections and operated by Amp, is expected to open in early 2026.

The greenfield project will be Waste Connections’ first MRF in Colorado. It will be able to process up to 62,000 tons of single-stream recycling annually, the companies said in a news release.

The facility will use a Amp One system, a custom building plan the company describes as a “zero-manual-sort” MRF that can be configured and optimized for various materials. The operational flexibility is meant to help Waste Connections adapt to changes in Colorado’s recycling market.

Colorado’s waste hauling landscape is particularly competitive, the companies say, and Waste Connections sees a focus on AI-powered sorting technology as an important strategy to continue to grow the business in the region.

The state is also poised to begin implementing an extended producer responsibility program for packaging, which the companies expect will influence recycling markets in the region. The new MRF will allow Waste Connections to adjust its scale and sorting capabilities as those changes take effect, said Mark Ceresa, the company’s division vice president.

“We just know that there's going to be a bigger market for recycling going forward because of EPR, and we'd like to meet that demand,” he said. “We feel as though the entire state of Colorado is going to need these types of facilities to be online, and we want to get out ahead of it.”

Upcoming EPR deadlines include a state recycling system plan expected in February 2025, which could give companies like Waste Connections insight into market changes. Plan implementation is expected in January 2026.

The MRF will be equipped with Amp’s equipment, which uses AI not only to sort materials but to monitor material purity and adjust efficiency. The facility will also be designed so the system can automatically detect problems such as jams or “drops in performance,” then automatically adjust equipment in real time, Amp said in a news release.

The Commerce City MRF will process recycling volumes across the state, including from major cities north and south of the Denver area such as Colorado Springs and Fort Collins, Ceresa said. Waste Connections currently sends the volumes it collects to third-party MRFs in the region.

Waste Connections already uses about 50 of Amp’s AI-powered robotic systems throughout North America, which the companies say make it the largest AI MRF operator in the sector. The Commerce City MRF will be the first time the companies have partnered on building an entire MRF, said Amp CEO Tim Stuart.

The agreement for Amp to design and operate the Commerce City facility was an ideal scenario for Waste Connections, Ceresa said. “Having them run the facility as well as provide the equipment, gave us the ability to de-risk what we were walking into with this MRF,” he said. 

Stuart said it’s also an opportunity for Amp to identify other areas in the competitive Colorado market where it can offer its equipment and services. 

“Being more efficient obviously makes you more competitive, makes you look to grow your business,” he said.

The company has recently built Amp One facilities in Cleveland, Ohio; Portsmouth, Virginia, and Pitt County, North Carolina. In the near future, “we have a lot in the pipeline on Amp One, both with Connections and some other companies,” he said.

Waste Connections is working on expanding its overall recycling efforts across the company and has a goal to increase recycling by 50% from a 2018 baseline.

Recycling project collaborators didn’t just talk the talk at the Packaging Recycling Summit, they walked the walk and shared tangible results from new and ongoing initiatives to transform how packaging is managed.

Teamwork across the supply chain was again an important theme at the June 2025 event, especially as various players grapple with what impact to expect from new state laws regarding extended producer responsibility for packaging.

These were some of the collaborative projects spotlighted at PRS in Irving, Texas.

Removing grey area

Speakers repeatedly referenced projects that harness artificial intelligence and data, such as a partnership between investment group Closed Loop Partners and waste analytics company Greyparrot. 

Closed Loop works toward circular economy solutions with partners across academia, nonprofits and the corporate world “because we know that one organization cannot drive this important transition alone,” said Kristin Taylor, director of catalytic capital.

The firm invested in Greyparrot in 2022 when it recognized the platform’s potential to recover more polyethylene and polypropylene, and “really for the entire recycling ecosystem,” Taylor said. It’s a tool enabling brand owners to gather data — such as collection data from MRFs — to better understand what happens to their packaging across waste and recycling supply chains and where improvements can be made.

This waste intelligence platform is the first dedicated to brands, said Gaspard Duthilleul, COO at Greyparrot. “We can track this packaging through the sorting centers and the recycling centers. We can see which ones are being sorted correctly and which ones are not.”

The system can recognize up to 111 different waste categories. It’s granular enough to identify individual brands’ packaging, such as which bottles are from the Coca-Cola Co. and which are from PepsiCo. The information is entered into the AI-driven database. “It’s a bit like Google Photo for waste, for innovation,” Duthilleul said of the system, explaining that users can create data filters to gain specific insights to inform changes.

Greyparrot’s system allows users to “unlock a new, unprecedented level of data to improve bale quality, to reduce residue, to reduce waste to landfill, and to provide actionable data that can really make the entire system more efficient, more resilient and more profitable,” Taylor said.

Pouch pick-ups

Baby food brand Once Upon a Farm sells many of its fruit and vegetable blends in flexible plastic pouches, which provide consumer convenience as well as the ability to use high-pressure pasteurization for processing without using heat. But pouches present recycling challenges.

The B Corp is working to improve its packaging in several ways, said Madeline Tien, a sustainability specialist at Once Upon a Farm. This includes design and materials management, such as using less plastic in each pouch and transitioning from a multilayer design to monomaterial polyethylene to improve recyclablility. The brand also leans on its partnerships to solve sticking points, Tien said, pointing to the collaboration with specialty recycling service provider Recyclops.

The partners developed a pouch mail-back program for consumers, which last year resulted in the collection of 120,000 pouches. Recyclops bales the collected pouches and ships them to a third-party processor that deals with hard-to-recycle plastics; the processor transforms the items into plastic pellets for reuse in new products.

Recyclops offers an alternative collection method for consumers who are skeptical about whether the items they put in curbside bins actually get recycled, said CEO Ryan Smith. “It’s a complex supply chain,” he said, noting hurdles such as few end markets for hard-to-recycle plastics. Recyclops’ work also involves connecting processors with brands to develop solutions for these hurdles and identify novel end markets.

“I really can't stress enough how valuable it is for us to be able to have this level of transparency into the process,” Tien said. “Collaboration, especially for us, really unlocks opportunities that wouldn't otherwise be available to us as an emerging brand.”

The partners acknowledge mail-back programs aren’t the sole recycling solution; they’re one option in the broader ecosystem and currently provide a testing ground for more comprehensive solutions down the road. The long-term hope is for these flexible plastic pouches eventually to be accepted curbside, with the take-back program serving as a stepping stone to get there, Smith said. “The beginning isn't the end. Just because we start a program in a certain way doesn't mean that's how the program needs to end.”

Breaking through beverage barriers

Improving beverage container recovery rates, especially to meet growing demand for postconsumer recycled material, also takes collaborative work, according to PRS speakers.

The goal is to “get a lot more [quality] material out of homes,” said Megan Daum, vice president of sustainability at American Beverage. “We need to make sure that we're linking that quality with MRFs that are upgraded.” 

But American infrastructure is lacking, and that presents a major barrier to achieving goals, said JD Ambati, founder and CEO of AI-enabled recycling technology company EverestLabs. While adding optical sorters has changed the MRF game over the last five years, such technologies have a long way to go before they’re considered mainstream and markedly move the needle on improving container recovery rates, according to speakers.

“In the world of processing, we had more innovations between the 1900s to 1990s than we had since the 1990s until now,” Ambati said. “I have a lot of respect for our [MRF] operators, because this is a very critical infrastructure that is running with very little modern technology.”

Speakers noted that grants remain available for equipment upgrades through groups like The Recycling Partnership. The Can Manufacturers Institute also offers funding opportunities for companies that wish to upgrade to modern technologies, such as robots.

“I'll fund anything that gets me additional cans captured and gets some upside” in terms of revenue sharing, said CMI President Scott Breen, noting his group’s financial assistance for technologies at waste and recycling services companies Caglia Environmental in California and LRS in the Chicago area.

“Every quarter, I get a very lovely email with my performance report, how many cans each day that robot captured, and my thousands of dollars in a check. And then I go and search for other opportunities” to invest that money, he said.

Ambati explained that sorting robots also can be adjusted to boost recovery of materials other than cans, such as plastics. Daum, looking at the broader industry implications of specific infrastructure investments, noted that “we love all our containers” and “the rising tide floats all boats.”

Beyond equipment investments, beverage brands and packaging manufacturers report working together to improve container capture rates by supporting the introduction of more bottle bills and navigating emerging EPR laws.

“We are working together because [whether] PET, glass [or] aluminum, we all need our material back so we can meet recycled content mandates. But also so we have domestic, resilient supply chains, and so we can make packaging that has a lower carbon footprint,” Breen said.

Ramping up research

The Center for Plastics, Paper and Hybrid Packaging just launched thanks to a grant from the National Science Foundation. It’s set up to be a research center that helps to drive packaging recycling innovation and develop skilled workforces by bringing together industry and academia.

“The current administration approving a federally funded and subsidized center in the field of package sustainability is a little bit of a surprise,” said Thomas Osip, industry liaison officer at C3PS. “It’s pretty exciting.”

The goal is to engage not just packaging experts, but also those in fields such as AI, regulatory science and social sciences to collaboratively “work on projects that might be difficult for an individual company, or maybe a little bit risky for them, to tackle alone,” Osip said. The expected output is technologies that enable recycling, as well as solutions for recyclable and compostable packaging. 

Member companies will pay a fee to be part of the program. At launch, the initiative’s participants include eight universities, two NGOs and 10 companies, Osip said. Michigan State University is the site hub for the network. C3PS plans to build out its website and operations in the coming months. 

C3PS asks industry participants their needs, existing problems and what they wish they currently had but don’t, and then it brings those ideas to researchers who develop project proposals. Some specific areas of interest right now are biodegradable barriers and scalable PFAS-free packaging solutions.

“What's cool about it is industry identifies the needs, industry selects the projects and ... they vote, rank and decide which projects go forward,” Osip said. “Projects get periodic updates. They get results.”