A research team at Stony Brook University is developing artificial intelligence tools to improve recycling processes and reduce contamination in waste streams. The project, funded by the university’s AI Innovation Seed Grant, uses video footage and advanced AI algorithms to automate the analysis of recycling materials.
According to estimates from the Environmental Protection Agency, about 75% of waste generated in the United States is recyclable, but only 35% actually gets recycled. This leads to approximately 68 million tons of recyclables ending up in landfills or incinerators each year. Contamination—when non-recyclable items are mixed with recyclables—occurs at a rate of 25%, resulting in millions of tons being rejected and sent to landfills.
The Stony Brook research team is working on an AI-assisted system that employs sensors and machine learning algorithms to identify, track, and count waste materials as they move through real recycling facilities. The goal is to create a smarter and more cost-effective approach for sustainable waste management.
“We’re not just building tools in isolation, we’re collecting data at multiple stages of the sorting process, engaging with recycling workers to understand the pain points, and using those insights to help them work faster, safer, and with greater insight,” said Ruwen Qin, associate professor in the Department of Civil Engineering.
The project involves collaboration with municipalities and the Waste Data and Analysis Center within the Department of Technology, AI, and Society. This center receives funding from the New York State Department of Environmental Conservation (NYSDEC). The researchers are gathering high-resolution video data from several stages of sorting at local materials recovery facilities (MRFs) on Long Island.
“We’re not just studying the problem. We’re building tools that can make a measurable difference,” said Qin.
The full story by Ankita Nagpal can be found on the AI Innovation Institute website.