<![CDATA[The increasing global challenges of waste accumulation, energy transition, and climate resilience have positioned Waste-to-Energy (WtE) systems as an important pathway for sustainable resource recovery. However, the complexity and uncertainty of WtE processes require advanced modeling approaches capable of capturing nonlinear relationships and improving predictive accuracy. Machine Learning (ML) has emerged as a promising tool for optimizing WtE operations; however, existing studies remain fragmented and are largely focused on technical performance rather than broader conceptual, methodological, and sustainability dimensions. This study aims to systematically examine the development, trends, methodological characteristics, and research gaps in ML applications for WtE systems. A Systematic Literature Review (SLR) was conducted using the PRISMA framework by analyzing relevant publications indexed in Scopus from 2020 to 2025. From the initial search results, 42 studies were selected for qualitative thematic analysis using structured coding to identify technological domains, geographic patterns, theoretical foundations, and methodological approaches. The findings indicate that ML applications in WtE are predominantly focused on optimization, prediction, gasification, and energy recovery modeling. However, the field remains characterized by geographical concentration, limited theoretical integration, insufficient interdisciplinary perspectives, and inconsistent consideration of governance, policy, and socio-environmental factors. The review highlights that future ML-based WtE research should incorporate stronger theoretical frameworks, improve data diversity, enhance model transparency, and integrate sustainability-oriented assessments. In conclusion, ML provides significant opportunities for advancing WtE systems; however, its transformative potential depends on moving beyond algorithmic accuracy toward holistic, equitable, and context-sensitive sustainability solutions.]]>
Copyrights © 2026
