Grantee Research Project Results

AI-Driven Recycling Centers for Disadvantaged Communities: Improving Recovery, Safety, and Costs

EPA Contract Number: 68HERC25C0006
Title: AI-Driven Recycling Centers for Disadvantaged Communities: Improving Recovery, Safety, and Costs
Investigators: Lamuraglia, Jack
Small Business: KLAW Industries LLC
EPA Contact: Richards, April
Phase: II
Project Period: October 18, 2024 through October 17, 2026
Project Amount: $400,000
RFA: Small Business Innovation Research (SBIR) - Phase II (2025) Recipients Lists
Research Category: SBIR – Sustainable Materials Management

Description:

Creating the Foundation for Modern Recycling Facilities The processing technology used in the recycling industry has remained unchanged since the 1980s, relying on manual labor and rudimentary machinery to solve our waste crisis. For the past 40 years, municipalities have shelled out egregious tipping fees for ineffective and dangerous sorting practices. Cities across the U.S. need a new recycling processing system to correct the backward economic model plaguing the sector. Our project has developed a rapidly deployable, autonomous robotic sorting system to build the foundation for the next era of recycling facilities. Rather than relying on slow-moving waste haulers, KLAW Industries has developed, deployed, and operated this technology to reduce municipal tipping fees by 13%. With an autonomous sorting system to match the modern waste stream, we have demonstrated our ability to reduce product contamination to 0.5%, lower capital costs by 58%, and decrease excess hauling by 94%. Technical Feasibility & Comparative Performance During Phase I, we demonstrated our system's accuracy of 90% and validated our 13% cost savings to municipalities by processing single-stream recycling directly from the City of Binghamton in pilot operations. KLAW Industries' researchers have made breakthrough innovations in feature extraction and convolutional neural networks during Phase I to make rapidly deployable autonomous sorting possible at a local level, which we have demonstrated on over 6 tons of recyclables. During Phase II, we will commercialize the technology and deploy a full-scale prototype to process recyclables in regular operations. We will continue our partnership with the City, bringing cost-effective recycling to the disadvantaged community of the East Side of Binghamton. Our target market is cities and commissioners of public works who face the unpleasant dilemma between paying more for recycling or saving money, time, and headaches by dropping the program, as many low-income towns and villages already have. Market Size & Performance Benefits & Environmental Benefits With a total available market (TAM) of $37.8B in the U.S., we will phase out diesel[1]powered hauling enterprises and usher in technology companies focused on long-term material recovery. The recycling sector injury rate is 3.2 per 100 full-time-equivalent workers. Autonomous sorting protects critical workers by moving away from conveyor belts and toward computer terminals. With an increased sorting accuracy of 90% and a 13% lower tipping fee validated during Phase I, we will begin sorting, classifying, and reusing waste locally to achieve a recycling rate of 50% by 2030.

SBIR Phase I:

AI-Driven Recycling Centers for Disadvantaged Communities: Improving Recovery, Safety, and Costs  | Final Report