Digital twins and AI redefine urban waste management worldwide

A surge in scientific research is reshaping the future of sustainable waste management through the integration of Artificial Intelligence (AI) and Digital Twins (DT), promising smarter cities, cleaner environments, and more circular economies. This transformation is captured in a sweeping bibliometric and thematic review titled “Artificial Intelligence and Digital Twins for Sustainable Waste Management: A Bibliometric and Thematic Review”, published in Applied Sciences.

The study synthesizes trends from 570 peer-reviewed publications indexed in Scopus, covering research outputs between 2015 and 2025. 

How are AI and digital twins reshaping waste management research?

According to the study, AI and DT technologies are at the center of a rapidly evolving scientific field aimed at tackling environmental degradation caused by mounting global waste. The researchers identified three major thematic clusters through VOSviewer-based bibliometric mapping: (1) predictive and environmental modeling, (2) sustainability and circular economy optimization, and (3) human-environment interaction and monitoring.

The first cluster, driven by keywords like "prediction," "machine learning," and "artificial neural networks," highlights AI’s central role in forecasting waste generation, landfill emissions, and organic matter decomposition. The second cluster, anchored in terms such as "digital twin," "circular economy," and "recycling", signals growing emphasis on integrating digital simulation into low-carbon waste treatment strategies. The third cluster, themed around environmental quality and health, is smaller but focuses on public risk assessment via real-time data systems like sensor networks and geospatial analytics.

Keyword co-occurrence analysis confirmed "prediction" as the most connected term, reinforcing AI’s dominant function in managing and forecasting waste. Other top-ranking keywords included "recycling," "waste management," and "adaptation strategies," reflecting the interdisciplinary scope of the field. This diverse keyword architecture reveals a convergence of AI technologies with practical waste recovery strategies, embedded within sustainability frameworks.

Who is driving the research and what are they studying?

The review found that while waste management remains a global concern, research activity is geographically concentrated. Australia leads the field in volume, with 39 publications and the highest collaboration metric (total link strength of 23), followed by Saudi Arabia, Iran, and the United States. Countries like Hungary and the Netherlands, though lower in publication count, show disproportionately high academic impact, as measured by citations per article.

Publication venues further validate the sector’s multi-domain orientation. Leading journals include Waste Management, Journal of Cleaner Production, and Science of the Total Environment, all reflecting a blend of technical engineering and sustainability research. These journals also dominate in citations, serving as intellectual anchors for emerging waste-tech paradigms.

Beyond bibliometric mapping, the researchers performed a thematic evaluation of the 20 most-cited articles. This analysis revealed six prominent research domains:

1. Urban Waste Forecasting – Researchers employed machine learning models, such as Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Recurrent Neural Networks (RNNs), to predict waste generation in cities like Ho Chi Minh and regions like Canada.
2. Recycled Materials Optimization – AI was used to simulate compressive strength, carbonation depth, and durability of sustainable construction materials like recycled aggregate concrete.
3. Circular Economy Digitalization – Studies highlighted Digital Twin applications in metallurgical processes to optimize resource use and enable near real-time decision-making.
4. Electronic Waste Monitoring – Machine learning models like Random Forests helped map contamination hotspots in heavy-metal-polluted soils.
5. Deep Learning for Environmental Health – Convolutional Neural Networks (CNNs) were deployed to detect water pollution in agricultural irrigation systems.
6. Industrial Integration of AI and DT – Frameworks such as CRISP-DM enabled industrial-scale predictive modeling for waste valorization and byproduct management.

The study illustrates a temporal evolution in research focus. From 2016 to 2018, AI applications were largely exploratory, focusing on individual forecasting tasks. Between 2019 and 2021, the emphasis shifted toward material design and optimization. By 2023, there was a marked rise in systems-oriented approaches integrating DTs and industrial governance.

What challenges and opportunities lie ahead?

Despite significant progress, the study highlights critical gaps. The integration of AI and DT into actual waste management systems remains uneven across regions and sectors. Many of the top-cited studies are based on simulations, with limited empirical validation in real-world waste management contexts. Furthermore, the research remains skewed toward engineering and environmental science domains, with minimal contributions from policy, social sciences, or governance frameworks.

The researchers call for a shift from reactive to anticipatory waste systems that continuously adapt using AI-powered simulations and real-time data. They advocate the concept of “anticipatory circularity,” where waste is not merely collected and processed, but preemptively optimized and revalorized.

To operationalize these insights, the study recommends:

• Empirical Deployment: Pilot studies of AI and DT applications in live urban environments and industrial chains to assess cost-effectiveness and scalability.
• Policy Integration: Aligning digital tools with circular economy regulations and governance structures.
• Algorithm Comparison: Benchmarking performance across AI models for different waste types (e.g., municipal, industrial, hazardous).
• Ethical and Workforce Considerations: Addressing challenges in data governance, user acceptance, and skill development.
• Inclusion of Emerging Technologies: Exploring the roles of blockchain, edge computing, and generative AI in intelligent waste systems.

The future of sustainable waste management depends on a systemic redesign, fusing predictive intelligence with environmental monitoring, cross-sector coordination, and circular economic logic.