Conventional digital recycling processes endure from important useful resource loss on account of insufficient materials separation and identification capabilities, limiting materials restoration. We current A.R.I.S. (Automated Recycling Identification System), a low-cost, moveable sorter for shredded e-waste that addresses this effectivity hole. The system employs a YOLOx mannequin to categorise metals, plastics, and circuit boards in actual time, attaining low inference latency with excessive detection accuracy. Experimental analysis yielded 90% general precision, 82.2% imply common precision (mAP), and 84% sortation purity. By integrating deep studying with established sorting strategies, A.R.I.S. enhances materials restoration effectivity and lowers obstacles to superior recycling adoption. This work enhances broader initiatives in extending product life cycles, supporting trade-in and recycling packages, and decreasing environmental affect throughout the availability chain.
