A significant challenge faced by many MRFs is the reliance on manual waste separation, which is less efficient and limits recovery rates. Modernization with automated systems and advanced AI visual technology is essential to meet EU recycling objectives for 2035. Additionally, proper management of organic waste, stronger citizen participation in selective collection, and investment in new infrastructure are critical for advancing Spain toward a sustainable circular economy.

MRFs are challenging environments for equipment deployment. The Pixelabs team had to ensure that AI-Waste Engine could operate effectively in such conditions. Deep learning algorithms were implemented to classify waste based on material typology and composition, with real-time data displayed on a dashboard.

AI-Waste Engine, combined with next-generation computer vision technology, identifies various materials in real time, including PET, PE, PAPER, BRIK, PP, and METAL. Key challenges included the deployment of camera points and subsequent implementation in the plant.

The technology utilizes a precision camera system designed for demanding MRF environments, capable of recognizing and distinguishing waste by material type and composition. Its unique value proposition lies in the use of color and hyperspectral cameras, molecular analysis, processing dense waste streams, and predicting visually similar waste. The automation and agility of the solution significantly enhance the overall operation, reducing reliance on manual labor and increasing classification accuracy.

The implementation of AI-Waste Engine led to measurable improvements in the MRF’s operations. The most significant impacts include reduced operational time, higher classification accuracy, and cost savings. By enhancing processes, infrastructure, and operational efficiency, the solution contributes directly to more effective waste management and supports the broader goals of the circular economy.