Artificial Intelligence for Sustainable Environmental Management: Integrating Stormwater and Wastewater Systems

Abstract

Sustainable environmental management demands a holistic, systems-thinking approach that moves beyond the operational siloes separating stormwater and waxstewater governance. Urban water systems — encompassing stormwater conveyance, combined sewer networks, and centralised wastewater treatment — are intrinsically interconnected, yet historically managed by distinct engineering disciplines, institutional actors, and regulatory frameworks. This institutional fragmentation imposes substantial environmental, economic, and social costs: suboptimal infrastructure investment, missed opportunities for resource recovery, elevated receiving water pollution loads, and inequitable distribution of flood risk and service quality. Artificial intelligence (AI) offers the integrative analytical capacity to bridge these siloes, enabling holistic water system optimisation grounded in sustainability principles. This article presents a comprehensive review of 103 peer-reviewed studies published between 2018 and 2025, examining how machine learning, deep learning, reinforcement learning, multi-agent systems, life cycle assessment-informed AI, and circular economy-oriented decision support tools are transforming the integrated management of stormwater and wastewater systems. We evaluate AI applications across six sustainability dimensions: resource recovery maximisation, energy self-sufficiency, ecological quality protection, social equity and environmental justice, circular economy integration, and governance and institutional capacity. Key findings reveal that AI-optimised resource recovery — including nutrient recapture, biogas generation, water reuse, and heat recovery — can convert WWTPs from net energy consumers to net energy producers in 34–58% of case study contexts, while AI-driven green-grey infrastructure integration reduces combined sewer overflow volumes by 38–61% in reviewed deployments.