Irena FATA - THE ROLE OF BIG DATA ANALYTICS IN THE CIRCULAR ECONOMY

Dissertation Title: THE ROLE OF BIG DATA ANALYTICS IN THE CIRCULAR ECONOMY

Author: Irena FATA

Institution: University of Tirana, Faculty of Economics, Department of Statistics and Applied Informatics
Field of study: Applied Statistics and Informatics / Applied Statistics and Mathematics in Economics
Publication date: 08.07.2026
The dissertation is published in Albanian.

© Copyright: Irena FATA

Published by the University of Tirana
Based on legal acts, regulations and policies of the UT

👉 Click here to view the full dissertation (PDF)

Abstract

This dissertation examines the intersection between Circular Economy (CE) and Big Data Analytics (BDA) in an integrated manner, aiming to contribute to the analytical understanding of the role of data in assessing and interpreting the circular transition. Despite the significant increase in scientific interest in this field, the contemporary literature is characterized by conceptual fragmentation, methodological heterogeneity and a lack of consolidated empirical evidence. In this context, the study addresses the gap between the analytical potential of BDA and the real results of circular performance.

The dissertation relies on an integrative methodological approach, which combines bibliometric analysis, semantic analysis through topic modeling techniques and empirical statistical analysis on multidimensional panel data at the country level. Bibliometric and semantic analysis are used to identify the scientific structure and conceptual content of the EQ-BDA field, while the integrative empirical analysis goes beyond a purely typological approach. In this context, it includes OLS regression on panel data with cluster-robust standard errors, comparing four alternative specifications (M_VAR, M_KL, M1, M2), in order to assess the relative contribution of structural and digital factors, as well as testing the role of digitalization as a proxy indicator of BDA. In addition, the dissertation also includes predictive modeling based on Machine Learning (XGBoost), to quantify the trajectory of urban waste generation and to evaluate alternative scenarios of circular policies.

The results show that the transition towards a circular economy is characterized by different structural configurations and non-linear trajectories, even in contexts with formally similar policy frameworks. The study emphasizes the role of Big Data Analytics as a mediating analytical mechanism that supports the interpretation of these configurations, without assuming universal transition models. The econometric analysis highlights that waste generation per capita and GDP per capita are the main explanatory factors, while digital capacity does not show a measurable direct effect, empirically confirming the mediating role of BDA. The predictive analysis shows a structural stagnation in the baseline scenario and a cumulative reduction of up to 6,971 tons by 2030 in the aggressive intervention scenario. The contribution of the dissertation lies in the theoretical, methodological and empirical integration of the EQ-BDA nexus, providing an analytical framework for understanding circular performance based on evidence.

Field: Circular Economy and Big Data Analytics in the Context of Sustainable Development

Keywords: Circular Economy; Big Data Analytics; Waste Management; Circular Performance; Empirical Analysis; Sustainable Transition; Environmental Indicators; OLS Panel Regression; Machine Learning; XGBoost; Time Series Forecasting; Policy Scenarios

Abstract

This dissertation examines, in an integrated manner, the interconnection between the Circular Economy (CE) and Big Data Analytics (BDA), aiming to contribute to the analytical understanding of the role of data in assessing and interpreting the circular transition. Despite the rapidly growing scientific interest in this field, the contemporary literature remains characterized by conceptual fragmentation, methodological heterogeneity, and a limited consolidation of empirical evidence. Within this context, the study addresses the persistent gap between the analytical potential of Big Data Analytics and the observed empirical outcomes of circular performance.

The dissertation relies on an integrative methodological approach that combines bibliometric analysis, semantic analysis using topic modeling techniques, and empirical statistical analysis of multidimensional panel data at the country level. Bibliometric and semantic analysis are used to identify the scientific structure and conceptual content of the EQ-BDA field, while the integrative empirical analysis goes beyond a purely typological approach. In this context, it includes OLS regression on panel data with cluster-robust standard errors, comparing four alternative specifications (M_VAR, M_KL, M1, M2) to assess the relative contribution of structural and digital factors, and to test the role of digitalization as a proxy indicator of BDA. In addition, the dissertation includes predictive modeling using Machine Learning (XGBoost) to quantify the trajectory of urban waste generation and evaluate alternative circular policy scenarios.

The findings indicate that the transition towards a circular economy is characterized by diverse structural configurations and non-linear trajectories, even across contexts with formally similar policy frameworks. Econometric results identify waste generation per capita and GDP per capita as the primary structural determinants of landfill dependency, while the digitalization variable yields no statistically significant direct effect, empirical confirmation of BDA's intermediary rather than causal role. Forecasting analysis reveals structural stagnation under the business-as-usual scenario (11,200–12,500 tons/year) and a cumulative reduction of up to 6,971 tons by 2030 under the aggressive intervention scenario (−4%/year). The study highlights the role of Big Data Analytics as an intermediary analytical mechanism that supports the interpretation of these configurations and the quantification of policy effects, without assuming universal transition models. The main contribution of the dissertation lies in the theoretical, methodological, and empirical integration of the CE–BDA nexus, offering an evidence-based analytical framework for understanding circular performance.

Field of study: Circular Economy and Big Data Analytics in the context of sustainable development

Keywords: Circular economy; Big Data Analytics; Waste management; Circular performance; OLS panel regression; Machine Learning; XGBoost; Medium-term forecasting; Policy scenarios; Empirical analysis; Sustainable transition; Environmental indicators