Digital Twin-Driven Intelligent Transformation of Solid Waste Treatment

Rapid global urbanization is driving a surge in solid waste generation, while conventional treatment systems face both environmental risks and operational uncertainty. Digital twins, which enable real-time mapping between physical assets and virtual spaces, offer a computable and verifiable route toward low-carbon, resource-efficient, and intelligent waste management when deeply integrated with the Internet of Things, big data, and artificial intelligence. This study develops a comprehensive review tracing the digital twins from static geometric mirroring to dynamic, cognitive co-symbiosis, and summarizes a multidimensional architecture spanning physical, virtual, data, service, and connectivity layers, together with coupling mechanisms involving IoT sensing, federated learning, multimodal big data, and large model agents. The study aims to provide a theoretical framework and methodological references for advancing digital twin-enabled solid waste valorization. Building on this framework, we examine recent progress in three representative application scenarios for solid waste treatment, and identify key technical bottlenecks, including heterogeneous data fusion, model generalization across facilities and contexts, and real-time computation under constrained resources. We highlight the need for standardization, uncertainty quantification, cybersecurity, and lifecycle evaluation to support reliable prediction, optimization, and decision-making in real operations. Finally, we discuss future directions such as edge intelligence and the integration of city-scale material and energy networks.