Greyparrot CEO: AI-enhanced waste management can unlock circular economy

14 Feb 2024 --- Greyparrot is accelerating the digitization and automation of the global waste sector by partnering with Netherlands-based recycling firm Bollegraaf. More efficient resource recovery and recycling will translate into a steadier stream of high-quality recyclates to meet increasing demand, according to Greyparrot’s co-founder and CEO Mikela Druckman. 

Along with the UK’s 2022 plastics tax, the EU’s imminent Packaging and Packaging Waste Regulation (PPWR) update is increasing demand for packaging made of recycled material.

“We’ve already heard from packaging producers that are feeling the pressure to balance competitive prices with recycled material content that meets upcoming regulatory requirements,” Druckman tells Packaging Insights. 

We sit down with Druckman to discuss the current waste management landscape and the challenges of integrating AI waste analytics into existing software and hardware systems at recycling facilities. She highlights the environmental and financial benefits of an AI-based waste management industry.

How would you assess the state of the current global waste management industry?
Druckman: The global waste management industry is doing critical work to tackle the waste crisis, which mitigates key climate threats like reliance on fossil fuels for packaging, pollution from landfills and biodiversity loss. The industry suffers from a lack of visibility of what is in the waste stream. Today, only 1% of waste gets monitored, and even in advanced economies, around 40% of material sorting is still done by hand. We see millions of tons of potentially-recoverable material lost to landfills and incinerators as a result. It’s both a financial and environmental issue — in the US, a lack of visibility and automation leads to US$80–120 billion being lost — and that’s just for plastics. 

Greyparrot integrates AI waste analytics into existing software and hardware systems at recycling facilities.At the same time, we know that recycling needs to work and can work as part of a connected waste hierarchy in countries like Germany and South Korea. Prevention and reuse are essential priorities, but recycling remains a crucial filter between consumption and pollution and a vital source of secondary resources for packaging producers. The skepticism surrounding recycling often results from viewing the process in isolation. Policy, product design and investment all mean that by the time material arrives at a recovery facility or recycling plant, it’s supposed to be there. 

How are you leveraging AI to support your technology portfolio?
Druckman: We’re applying AI to increase global recycling efficiency and save millions of tons of resources from landfills and incinerators. We collect waste data using the Greyparrot Analyzer. Our camera systems are mounted on moving conveyor belts in recovery facilities, and our AI reveals real-time insights on over 70 waste categories across seven layers of data, including material type, brand, SKU, financial value, potential GHG emissions and food-grade. 

Our integration technology makes waste intelligence accessible to the entire waste recovery ecosystem. Greyparrot Sync APIs (Application Programming Interface) enable us to collaborate with third-party hardware and software developers, who use Sync to embed Greyparrot AI in their machinery and facility management software. Packaging producers and FMCG brands already use this waste data to make decisions that help them comply with recycling regulations and improve recyclable packaging design.

On what basis will the MRFs and PRFs you are planning to equip with AI capabilities be selected?
Druckman: There are 100 Greyparrot Analyzers active in 50+ facilities across 14 countries. Our units are retrofittable, meaning any facility with a standard conveyor belt can use our technology without adjusting their existing infrastructure. We want to put waste intelligence in the hands of as many waste professionals as possible. Every facility’s material reveals new insight into product recyclability on a local, national and global scale. The more data we gather, the more we’ll be able to offer granular feedback on the post-consumption life of packaging and guide sustainable product design. Our partnership with Bollegraaf represents a significant step toward the widespread adoption of AI waste analytics for the global waste sector. As part of the non-exclusive commercial agreement, Bollegraaf will distribute Greyparrot Analyzers into new and existing facilities.

What are the challenges associated with integrating AI waste analytics into software and hardware systems?
Druckman: Many AI waste analytics platforms are built to work alongside specific hardware or to create a closed sorting system. That can mean that AI — however powerful or accurate — works in isolation from the rest of the sorting processes in the facility rather than using recognition data to “level-up” the entire facility. AI has an incredible amount of potential, which until now has often been limited to smart sorting arms.Greyparrot collects waste data using the Greyparrot Analyzer.

Collaboration has always been a priority at Greyparrot, and we’ve addressed the challenge of integration by building a hardware and software-agnostic waste intelligence platform. We developed Greyparrot Sync — our suite of APIs — to make Greyparrot insight as accessible to fellow waste sector innovators as possible. We’ve seen our open approach translate into industry-first tools. Sync integrations and more hands-on collaboration are behind automated deposit returns sorting in South Korea, through to the first “digital twin” software for recovery facilities and automated sampling cabins. Soon, we’ll see those integrations deliver facilities that are connected, monitored, validated and adjusted automatically.

What can you tell us about the reduced loss in financial value by operating waste sorting facilities with AI?
Druckman: It depends on the capacity of the plant and the types of material it processes. In a plant that can handle 30 tons of material an hour, savings can easily reach €1.5–2 million (US$1.6–2.1 million) every year. That’s extremely valuable in an industry with tight margins. Those savings and revenue boosts come in a number of forms. A complete picture of waste stream content enables waste professionals to monitor and increase the quality of the materials they send to their customers, including recycled packaging producers. It also empowers them to identify and recover valuable resources lost to residue lines. In 2023, we found that 26% of the material that global recovery facilities sent to landfills was recyclable plastic.

There’s a financial impact on packaging producers and brands, too. Our AI’s ability to accurately recognize brands and SKUs means we’re gathering some of the first large-scale insight into the post-consumption life of many products. With regulations like extended producer responsibility on the near horizon, it’s more valuable than ever for producers to understand how their products perform in recovery facilities to make improvements and avoid penalties.

The benefits aren’t solely financial. This technology is helping the industry make a measurable reduction in waste sent to landfills and incinerators, which in turn cuts CO2, methane, ethylene and nitrous oxide emissions. A more efficient waste and recycling sector provides a steady stream of recycled material that will, with the help of policy, reduce our reliance on virgin materials produced using fossil fuels.

By Natalie Schwertheim