Measuring the level of circular economy in cities in alignment with sustainable development goal 11: Integration of machine learning algorithms and analytical hierarchical-network processes. A case study in Seville

The Circular Economy (CE) plays a crucial role in both transforming cities into inclusive, resilient, sustainable communities and addressing Sustainable Development Goal 11 (SDG11). However, current methods for the evaluation of city circularity are often incomplete due to their lack of multi-dimensional and comprehensive models and processes that encompass the three core pillars of sustainability, social, economic, and environmental, and SDG 11. This study addresses these limitations, by enabling precise hierarchical and network-based assessments and providing a quantitative score of city circularity. This hybrid model incorporates Multi-Criteria Decision-Making (MCDM) methods structurally inspired by Analytic Hierarchy Process (AHP) and Analytic Network Process (ANP), and Machine Learning (ML) models (pre-trained transformers, and Ensemble Learning (EL) techniques). The model operates across four levels: metrics, indicators, sustainability dimensions, and “circularity based on SDG11” (main goal) and measures their links. Semantic assessment and text mining analyze 123 metrics from 59 identified Circular City Indicators (CCIs). An ideal circularity score of 19.57 (equivalent to 100%) is calculated based on the maximum normalized values for all metrics. The proposed model is applied to the current data of Seville (Spain) in 2023, yields a circularity score of 7.25 (37.05%) under the proposed framework. The framework identifies key metrics with the highest impact on the final score while also recognizing secondary metrics with less influence on Seville's circularity score. This framework is a comprehensive tool for measuring city circularity, enabling cross-city comparisons and temporal analyses, while also being a powerful and adaptable instrument for different city levels, helping policymakers develop sustainable strategies.