دانشگاه بیرجند مطالعات مدیریت توسعه سبز 2981-2402 مقالات آماده انتشار 2025 11 18 Application of Artificial Intelligence in Urban Environmental Management: A Structured Review in Three Areas of Air Quality, Waste, and Energy Consumption کاربرد هوش مصنوعی در مدیریت محیط زیست شهری: مرور ساختاریافته در سه حوزه کیفیت هوا، پسماند و مصرف انرژی 3781 10.22077/jgdms.2025.9710.1322 FA راحله والیزاده اردلان دانشجوی دکترای علوم و مهندسی محیط زیست دانشکده منابع طبیعی و محیط زیست دانشگاه بیرجند جواد ستوده دانشجوی دکترای علوم و مهندسی محیط زیست دانشکده منابع طبیعی و محیط زیست دانشگاه بیرجند 0009-0007-8649-7677 الهام یوسفی دانشیار گروه محیط زیست، دانشکده منابع طبیعی و محیط زیست، دانشگاه بیرجند Journal Article 2025 07 14 Introduction The rapid growth of urbanization in recent decades has posed numerous environmental challenges for cities. Increasing urban populations, excessive resource consumption, high volumes of waste generation, and intensifying air pollution are among the most critical urban environmental issues. In this context, Artificial Intelligence (AI) has emerged as a transformative technology offering unparalleled capabilities for monitoring, analyzing, and managing complex environmental problems. Techniques such as machine learning and deep learning enable the identification of hidden patterns in large datasets, prediction of environmental trends, and provision of optimal solutions for resource management. Given the significance of this topic, the present study adopts a structured review approach to systematically examine and categorize published research on the application of AI in three key domains: air quality management, waste management, and energy consumption optimization in urban environments. Methodology This study employs a narrative systematic review approach to analyze research published between 2017 and 2025 across reputable scientific databases including Web of Science, Scopus, IEEE Xplore, ScienceDirect, and Google Scholar. The literature search was conducted using a set of carefully selected keywords such as “Artificial Intelligence,” “Machine Learning,” “Urban Environmental Management,” “Smart Cities,” “Air Quality Prediction,” “Waste Management,” and “Energy Optimization.” Inclusion criteria were established to focus on studies that apply AI techniques specifically within the three main domains of interest: air quality management, waste management, and energy consumption optimization in urban environments. Only peer-reviewed journal articles published in English were considered. Conference proceedings, non-urban studies, and articles without full-text access were excluded to maintain the quality and relevance of the review. Initially, 356 abstracts were screened for relevance. Following this, 100 articles were selected for detailed review, and ultimately, 55 full-text articles were thoroughly analyzed. Key data extracted from these studies included research objectives, AI methodologies employed, application areas, principal findings, challenges encountered, and identified limitations. The extracted data were qualitatively synthesized to identify emerging trends, gaps, and opportunities. The review methodology also incorporated a structured data extraction process, where information was systematically recorded in a predefined spreadsheet to ensure consistency and comprehensiveness. This process involved critical appraisal of the methodological rigor and relevance of each study, enabling a balanced assessment of the state of the art in AI applications for urban environmental management. By integrating qualitative and quantitative insights from diverse studies, the review provides a comprehensive understanding of how AI techniques such as machine learning algorithms (e.g., LSTM, Random Forest, CNN) are utilized to model, predict, and optimize environmental indicators and urban resource management. This approach facilitates the identification of best practices and highlights areas requiring further research and technological development. Finding The systematic review revealed that machine learning and deep learning algorithms, particularly models such as LSTM, Random Forest, and CNN, play a significant role in monitoring and predicting environmental indicators, optimizing resources, and enhancing urban resilience. In the domain of air quality, machine learning models analyzing satellite, meteorological, and traffic data have substantially improved the accuracy of air pollution predictions (e.g., PM2.5, NO2). Studies indicate that leveraging secondary data and advanced models enhances classification and forecasting of air quality indices, providing effective tools for environmental decision-making. In waste management, computer vision and deep learning technologies have been utilized for identifying and separating recyclable materials, optimizing waste collection routes and schedules, and monitoring waste movement. These approaches have resulted in cost reductions, increased recycling rates, and decreased pollution. Regarding energy consumption management, AI analyzes energy usage data in buildings and cities to identify consumption patterns and offer optimization recommendations. Furthermore, machine learning models significantly aid in demand forecasting and managing smart energy grids, especially in integrating renewable energy sources. Additionally, predictive maintenance of equipment using AI prevents energy wastage and reduces additional costs. Discussion The findings of this study demonstrate that AI can markedly enhance the accuracy, efficiency, and speed of urban environmental monitoring and management. The use of advanced machine learning models facilitates the analysis of complex and large-scale data, supporting data-driven decision-making by urban managers. However, challenges such as dependence on high-quality data, implementation costs, model complexity, and limited interpretability of results must be addressed. Developing robust data infrastructures, improving explainable AI models, and fostering inter-organizational policy coordination are proposed solutions to overcome these challenges. Ultimately, integrating AI into urban environmental management plays a pivotal role in achieving sustainable development goals and shaping smart cities, thereby paving the way for more resilient and sustainable urban futures. با رشد شتابان شهرنشینی و تشدید چالش‌های زیست‌محیطی، بهره‌گیری از فناوری‌های نوین به‌ویژه هوش مصنوعی (AI) در مدیریت محیط زیست شهری به ضرورتی انکارناپذیر بدل شده است. این مطالعه با رویکرد مروری ساختاریافته، به تحلیل و طبقه‌بندی نظام‌مند مطالعات منتشرشده بین سال‌های ۲۰۱۷ تا ۲۰۲۵ پیرامون کاربرد هوش مصنوعی در سه حوزه کلیدی مدیریت کیفیت هوا، مدیریت پسماند، و بهینه‌سازی مصرف انرژی پرداخته است. جستجو در پایگاه‌های معتبر علمی و اعمال معیارهای دقیق ورود، منجر به انتخاب ۵۵ مقاله جهت تحلیل نهایی گردید. نتایج نشان می‌دهد که الگوریتم‌های یادگیری ماشین و یادگیری عمیق، به‌ویژه مدل‌هایی چون LSTM، Random Forest و CNN، توانسته‌اند در پایش و پیش‌بینی شاخص‌های زیست‌محیطی، بهینه‌سازی منابع، و ارتقای تاب‌آوری شهری عملکرد قابل توجهی از خود نشان دهند. در کنار مزایا، چالش‌هایی نظیر وابستگی به داده‌های باکیفیت، هزینه‌های پیاده‌سازی، پیچیدگی مدل‌ها و ضعف در تفسیرپذیری نیز شناسایی شد. در نهایت، با ارائه مجموعه‌ای از پیشنهادات در زمینه ارتقای زیرساخت‌های داده‌ای، توسعه مدل‌های تبیین‌پذیر و سیاست‌گذاری بین‌نهادی، این مقاله بر نقش محوری هوش مصنوعی در نیل به اهداف توسعه پایدار شهری و تحقق شهرهای هوشمند تأکید دارد.

https://gmd.birjand.ac.ir/article_3781_691ce372c433a094b0ddec93da6a7ac1.pdf