Integration of artificial intelligence in advanced oxidation processes for sustainable wastewater treatment: A bibliometric and scientometric analysis (2014–2025)
Reyhan Ata
Abstract
Artificial Intelligence (AI) has significantly advanced the efficiency, adaptability, and sustainability of Advanced Oxidation Processes (AOPs) in wastewater treatment. This study conducts a comprehensive bibliometric and scientometric analysis of 2933 publications from 2014 to 2025, retrieved from the Scopus database, to map the evolution of AI-AOP research. China leads with 839 publications, while Chemosphere is the most active journal, publishing 99 articles. The cumulative citation count reached 28,380 by 2025, reflecting growing academic interest. Beyond trend analysis, a novel conceptual framework is proposed to illustrate integration points between AI techniques and AOP mechanisms, including kinetic moeling, multi-objective optimization, pollutant-specific degradation, and real-time system control. The findings reveal a shift from empirical approaches to predictive and self-adaptive systems, utilizing models such as ANN, CNN, LSTM, and hybrid algorithms. These AI-driven strategies have achieved pollutant removal efficiencies exceeding 95 % in various applications. The integration of AI has facilitated enhanced process automation, control of operational parameters, development of novel materials, and energy efficiency improvements. This study highlights the transformative role of AI in shaping next-generation AOP systems and anticipates that future advancements in AI will drive sustainable innovation in wastewater treatment technologies. • Bibliometric and scientometric analysis of 2933 AI-AOP papers reveals global research trends (2014–2025). • A novel framework maps AI-AOP integration points in modeling, optimization, and real-time process control. • China leads with 839 papers; Chemosphere is top journal. Total citations reached 28380 by 2024. • Thematic shift observed from ANN-based models to deep learning and hybrid AI (e.g., ANN-GA, CNN, LSTM). • Strategic gaps identified, energy-efficient modeling, and industrial-scale AI-AOP applications.