This study provides a technology- and sustainability-driven assessment of the waste management innovation landscape by integrating social network analysis, natural language processing, patent clustering, life-cycle evaluation, and multi-criteria decision-making. A dataset of 46,905 patents retrieved from Lens.org was analyzed using a BERT-based semantic embedding model combined with network community detection and logistic life-cycle modeling, resulting in 16 statistically validated technological clusters (Davies-Bouldin optimal k = 16), spanning high-temperature thermochemical conversion and pyrolysis reactors, industrial incineration, smart waste monitoring and digital control systems, fluid handling, chemical and hydrometallurgical processes, biomass-to-biofuel fermentation, anaerobic digestion, polymer recycling, enzyme engineering, and construction waste valorization. Life-cycle assessment highlights heterogeneous trajectories, with mature solutions like thermochemical reactors and incineration systems coexisting with rapidly growing domains such as digital monitoring and organic waste valorization. Network analysis identifies institutional and industrial hubs orchestrating knowledge flows, while multi-criteria prioritization through TOPSIS ranks technological clusters based on their impact, diversification, collaboration depth, and sustainability potential. Results emphasize the strategic relevance of biotechnology-driven waste treatment and bio-based energy recovery in advancing circular economy objectives. The findings provide actionable insights for evidence-based innovation strategy, sustainable technology investment, and policy design, revealing how targeted R&D can accelerate both environmental and resource efficiency outcomes in modern waste management systems.
Keywords: Circular economy; Patent analytics; Social network analysis; TOPSIS; Technology life cycle; Transformer-based embedding; Waste management.
Copyright © 2026 Elsevier Ltd. All rights reserved.