AI Could Improve Sustainability Through Smart Recycling Sorting

0
84

Welcome to Thomas Insights — every day, we publish the latest news and analysis to keep our readers up to date on what’s happening in industry. Sign up here to get the day’s top stories delivered straight to your inbox.

Plastic recycling plant with piles of single-use plastic water bottles.

The U.S. is home to over 2,000 active landfills and produces more than 254 million tons of trash each year. Today, 90% of this waste is not recycled due to inefficient processes, rising costs, and a lack of education and awareness about how to recycle.

More than 50% of Americans admit to being confused about recycling and 90% say they would recycle more if only it were easier.

How Robots are Improving Recycling Practices

A potential answer to the nation’s trash challenge? The rise of artificially intelligent robots with arms that can sort and pick waste from conveyor belts. These robots are equipped with pincers or suction cups that can quickly sort trash, newest models are capable of making between 60-80 picks per minute. 

What better application of AI than the development of robots that can step up and assist humans with the most undesirable tasks? In Material Recovery Facilities (MRFs) whose purpose is to receive, separate, and prepare recyclable materials, several operators have commented on the difficulties they’ve had recruiting and retaining laborers. Waste Management, for example, employs between 2,500 and 3,000 sorter positions but recruits approximately 15,000 people per year from staffing agencies to keep them filled. High turnover slows productivity, drives up costs, and impacts employee morale.

Addressing labor shortages is a key reason for the uptick in advanced automation and robotics being developed for MRFs by companies including AMP Robotics, Bulk Handling Systems, ZenRobotics, and Machinex. But there’s a whole host of other benefits that come with the adoption of this technology.

Improving Worker Safety

Sorting through waste presents significant safety concerns for employees. According to the University of Illinois School of Public Health, recycling workers are two times more likely to be injured at work, with fatality rates. MRF robots negate the need for workers to be exposed to hazardous situations, leaving them free to take on higher-skilled jobs such as management or supervision roles within the facility.

Using “Visioning” to Improve Quality Control

Robots are becoming increasingly effective with the ability to respond to multiple variables and provide good quality control, which results in less contamination in the sorted waste.

Unlike typical industrial robots which are simply programmed to perform repetitive tasks, AI-powered robots for MRFs are much more advanced, using “visioning” to enhance performance.

The most sophisticated systems collect data with 3D cameras that can identify types of material and distinguish between colors, shapes, sizes, and textures, enabling them to carry out positive sorting (extracting the valuable items rather than removing contaminated ones). The Distributed Robotics Lab at MIT, for example, is making its robots more tactile by developing sensors so they can determine items with their fingers. They’ve also developed a gripper that can pick up objects 120 times its weight.

Reducing Costs (In the Long Run)

Recycling costs are continuing to soar, especially as a result of China’s 2018 National Sword policy which placed severe restrictions on the import of contaminated waste. With the U.S. unable to process the vast amounts of garbage they are now being forced to handle, more waste than ever before is being sent to landfills or incinerators.

Overseen by MRF workers, robots are helping to address this sudden increase in waste, working continuously and consistently.

Contributing to a More Sustainable Society

Analysis from Verisk Maplecroft revealed that while the U.S. represents 4% of the world’s population, it produces 12% of municipal solid waste, making it the one developed nation whose waste generation surpasses its capacity to recycle. Globally, more than 2.1 billion metric tons of waste are generated every year and just 16% of this is recycled. Instead of expecting humans to get recycling right, we could start putting our faith in the sorting ability of AI.

What’s Delaying Mass Adoption of AI in Material Recovery Facilities?

Implementing this technology is not without its challenges. Many MRFs have mentioned that the suction grippers used on robot sorters require frequent replacement due to them easily wearing out. Manufacturers are working to produce more resilient suction grippers that can withstand harsh environments. 

Additionally, existing AI technology struggles to identify lightweight materials that are often used in product packaging. Refining the technology will ensure lighter materials are not scooped up alongside other items.

Even establishing strong and reliable interconnectivity between automated machinery and MRF employees is a tricky hurdle to overcome. The upfront cost of investing in automation is a deterrent for MRFs, particularly smaller businesses, even if it does promise long-term ROI.

To date, venture capitalists have not been rushing to invest in this technology, which perhaps explains why automation in MRFs has taken time compared with other industries. Ultimately, as the technology matures, MRFs and investors have the opportunity to clean up in terms of both profitability and sustainability.

Image Credit: Alba_alioth / Shutterstock

Strategizing Your Supply Chain Like a Game of BattleshipNext Story »

More from Industry Trends



Original Source

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.