From Silos to Synergy: Transforming Urban Waste Efficiency Through Collaborative Multi-Agent Systems: A Scoping Review
Year: 2026
Type: conference-paper
Study design: Systematic Review
Abstract: This research presents a scoping review of collaborative multi-agent systems (MAS) for sustainable waste management, addressing the global challenges of urbanization and population growth. Using the PRISMA-ScR methodology with a four-stage systematic analysis process, 22 relevant papers were retrieved and analyzed from the Scopus, IEEE Explore, and Web of Science databases. The results identify several dominant MAS paradigms, including multi-agent reinforcement learning (MARL), bio-inspired models such as Ant Colony and Artificial Bee Colony algorithms, and Agent-Based Modelling (ABM) integrated with Geographic Information Systems (GIS). Key implementation challenges include scalability in dynamic environments, communication overhead, infrastructure constraints, and technical complexities in detecting waste in partially observable settings. Despite these barriers, the review highlights significant efficiency gains: MAS-driven approaches can reduce fuel consumption by up to 25 percent, collection time by 40 percent, and bin overflow events by up to 95 percent, respectively. Additionally, blockchain-integrated systems achieved a waste classification accuracy of 97.3 percent with low latency, while satellite-based architectures detected illegal dumping sites with 97.25 percent accuracy. Collaborative MAS is a vital tool for adaptive and sustainable smart city systems, emphasizing the importance of lightweight security protocols and practical urban implementation.
Language: English
Subfield: Waste Management and Disposal
Field: Environmental Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7220967487