Artificial Intelligence for Sustainable Energy Management in Smart Cities: A Review of Decision Support Systems and Applications

The increasing complexity of smart cities, characterized by fluctuating energy demand, distributed renewable integration, and dynamic urban processes, necessitates advanced decision-making tools that can operate under uncertainty. Artificial intelligence (AI)-driven decision support systems (DSS) have emerged as promising solutions, offering predictive and optimization capabilities to enhance sustainable urban energy management. This paper presents a comprehensive review of recent contributions (۲۰۲۵) that apply AI within DSS frameworks across diverse smart city domains. A total of fifteen peer-reviewed studies were analyzed, encompassing applications in smart grids, smart buildings, waste-to-energy systems, infrastructure, logistics, and urban informatics. The review identifies prediction and optimization as the unifying methodological features across these works. AI techniques such as machine learning, deep learning, reinforcement learning, fuzzy decision-making, and hybrid neuro-symbolic models are applied to forecast energy demand, renewable generation, waste flows, and supply chain dynamics, while simultaneously optimizing energy distribution, cost efficiency, system resilience, and policy alignment. Findings indicate that AI-driven DSS substantially improve forecasting accuracy and resource allocation, thereby advancing sustainable urban development. However, challenges remain in terms of scalability, data interoperability, model interpretability, and integration with socio-economic and regulatory frameworks. The review highlights these gaps and proposes directions for future research, including scalable and interoperable architectures, explainable AI, and frameworks for human-AI collaboration in participatory governance.

کلیدواژه ها:

Artificial Intelligence (AI) ، Decision Support Systems (DSS) ، Smart Cities ، Sustainable Energy Management ، Predictive Optimization