Managing municipal living plastic waste (MLPW) entails complex system-level challenges across collection, recycling, and treatment, necessitating simultaneous optimization of resource, environmental, and economic objectives. However, robust assessment is frequently impeded by data scarcity and measurement inaccuracies, which undermine the transferability of evaluation models. To address these limitations, a recent study published in Engineering by Ziyang Wang, Shen Yang, Junqi Wang, and Shi-Jie Cao proposes an artificial intelligence (AI)-enhanced evaluation framework to provide critical insights into city-scale MLPW management and strategies for optimizing its low-carbon and economic performance.

The study highlights that effective MLPW management is influenced by a complex interplay of physical material flows, spatial infrastructure, and socio-economic factors. These include waste composition, population density, economic activity, and specific end-of-life pathways such as incineration, landfilling, and recycling. The authors discuss how these factors affect life-cycle carbon emissions and process-based costs, which are crucial for evaluating the feasibility of transitioning to a circular plastics economy.

One key methodological advancement is the integration of machine learning algorithms to address data gaps and validate foundational city-level plastic data. The baseline material flows were derived directly from systematic field measurements and differential scanning calorimetry (DSC) characterization. However, potential biases were mitigated through an artificial neural network (ANN) model. The article emphasizes the importance of using multi-source covariates to conduct independent cross-checks, missing-data imputation, and explicit uncertainty propagation, ensuring the credibility of the subsequent environmental and economic assessments.

The authors delve into the effects of various intervention scenarios on MLPW carbon mitigation potentials. For example, the implementation of source reduction and bio-based material substitution can deliver robust near- to mid-term emissions reductions. The integration of high-quality recycling pathways, however, exerts the dominant mitigation effect. The optimal composite scenario yields a 96.3% reduction in annual greenhouse gas emissions by 2060 relative to the baseline.

The assessment also addresses the economic trade-offs and technological challenges associated with different MLPW treatment pathways. It suggests that mechanical recycling is currently more appropriate than chemical recycling for near-term implementation due to its superior cost-effectiveness and process maturity. The authors highlight that mechanical recycling achieves an emission intensity of approximately 108 kg CO₂-eq/t and generates an economic return of around 613.9 CNY/t. Cumulatively, the optimal trajectory achieves a reduction of 22.22 Mt CO₂-eq and generates economic benefits of approximately 197.7 billion CNY.

The study concludes with strategic recommendations for tailoring MLPW management pathways. The authors suggest that policymakers should treat source reduction and design for circularity as long-term constraints, rather than relying solely on high recycling rates which can dilute long-term climate performance. Future efforts should focus on prioritizing improvements in mechanical recycling infrastructure while advancing chemical recycling steadily through targeted demonstration projects.

This comprehensive framework provides highly transferable guidance for researchers and urban planners aiming to optimize MLPW governance under data-constrained conditions. Ultimately, it contributes quantitative evidence to support policy prioritization, facility location, and budget planning in global zero-waste city initiatives.

The paper “AI-Enhanced Assessment Framework for City-Scale Management of Municipal Living Plastic Waste Towards Zero-Waste Cities,” is authored by Ziyang Wang, Shen Yang, Junqi Wang, Shi-Jie Cao. Full text of the open access paper: https://doi.org/10.1016/j.eng.2026.03.009. For more information about Engineering, visit the website at https://www.sciencedirect.com/journal/engineering.