The rStream trailer unit is designed to process up to one ton of waste per hour, separating recyclables from trash in real time. The system identifies items on a conveyor belt and automatically redirects them into appropriate waste streams—recovering materials that would otherwise be lost in landfill-bound trash.
Unlike many conventional systems that only clean up recycling contamination, rStream also extracts recoverable items from general waste. This two-way sorting model helps reduce material loss and enhances overall recovery rates—critical benefits for organizations looking to improve sustainability metrics without scaling up infrastructure.
UMass Amherst provides a controlled yet diverse setting for deploying and refining the rStream system. With varied waste sources generated by students, staff, and faculty, the campus simulates real-world conditions that push the AI to handle a wide range of items—including food-contaminated packaging and non-recyclables.
The data collected through the pilot informs how the system handles sorting complexity, and it supports continual AI training to increase performance over time. The result is a more adaptive solution that’s well-suited to environments where standard industrial-scale recycling isn't practical.
One of the system’s key value propositions is its ability to generate actionable data. By identifying waste composition in detail, rStream enables institutions to collaborate with contractors and material buyers to support circular systems. At UMass, this data may eventually help convert recovered materials into new goods, such as apparel or merchandise for campus retail outlets.
This approach adds operational value by improving transparency, targeting higher-value materials, and aligning sustainability initiatives with measurable recovery outcomes.
rStream’s mobile system is built for smaller-scale operations, addressing the market gap between household recycling and large-scale sorting facilities. This flexibility positions it well for applications in universities, corporate campuses, hospitals, or event venues—anywhere with moderate but diverse waste output and a need for cost-effective material recovery.
By delivering a compact yet capable system, rStream provides a realistic recycling option for organizations without access to centralized sorting infrastructure, while maintaining high throughput and AI learning capacity.
