ARTIFICIAL INTELLIGENCE AND THE TRANSITION TO SMART RECYCLING: ADVANCES, CHALLENGES, AND OPPORTUNITIES

Abstract

Abstract- The growing generation of municipal solid waste (MSW), driven by urbanization and population growth, has intensified the need for innovative solutions to improve the efficiency and sustainability of recycling systems. In this context, artificial intelligence (AI) is emerging as a key technological enabler within the smart city paradigm. This study developed a systematic literature review (SLR) with the aim of analyzing the advances, challenges, and opportunities associated with the application of AI in the transition to smart recycling. The PICO approach methodology was used to filter the scientific literature in Scopus and Web of Science between 2020 and 2025, from which 35 relevant articles were selected for in-depth analysis. The results show that deep learning models such as convolutional neural networks (CNN), YOLO, EfficientNet, and hybrid models in automatic classification already break the 90% accuracy barrier. This advance renders traditional mechanical classification obsolete, transforming waste logistics into a key part of the circular economy that recovers previously discarded materials. However, high costs, the fragility of models in real environments, and the difficulty of scaling these solutions remain areas for further research. The future looks promising, with the trend pointing toward a convergence between edge computing, collaborative robotics, and blockchain. This technological mix promises a waste management ecosystem that is not only smart, but truly autonomous and sustainable.