Internet of Things and Machine Learning for Sustainable Development: A Systematic Review and Conceptual Framework

Abstract

The integration of the Internet of Things (IoT) and Machine Learning (ML) has emerged as a transformative technological paradigm for addressing global sustainability challenges through intelligent sensing, real-time data analytics, and autonomous decision-making. Although numerous studies have explored IoT-ML applications in domains such as healthcare, agriculture, smart cities, energy, transportation, manufacturing, and environmental monitoring, existing review articles largely provide descriptive summaries with limited critical analysis, comparative evaluation, and theoretical integration. This study presents a systematic literature review (SLR) to critically evaluate recent advances in IoT-ML technologies for sustainable development, identify research gaps, examine implementation challenges, and propose a novel conceptual framework to guide future research and practice. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines, with peer-reviewed publications retrieved from major scientific databases using predefined inclusion and exclusion criteria. The selected studies were critically synthesized through comparative analysis based on application domains, machine learning techniques, predictive performance, scalability, computational efficiency, energy consumption, and implementation cost. The findings indicate that the integration of IoT with advanced ML techniques-including Random Forest, Support Vector Machine, Artificial Neural Networks, Deep Learning, XGBoost, LightGBM, Edge AI, and TinyML-significantly enhances predictive analytics, intelligent decision-making, and resource optimization across diverse sustainability applications. However, persistent challenges related to interoperability, cyber security, privacy, computational complexity, infrastructure limitations, and policy readiness continue to hinder large-scale deployment, particularly in developing countries. As its principal contribution, this review proposes a Conceptual IoT-Machine Learning Framework for Sustainable Development that integrates IoT sensing infrastructure, communication technologies, edge-fog-cloud computing, intelligent analytics, decision-support systems, governance mechanisms, and sustainability outcomes within a unified architecture aligned with the United Nations Sustainable Development Goals (SDGs). The review also outlines emerging research directions, including Explainable Artificial Intelligence (XAI), Digital Twins, Federated Learning, Sustainable AI, Green AI, and next-generation AI-powered IoT architectures. The proposed framework provides valuable theoretical insights and practical guidance for researchers, policymakers, and industry practitioners in designing scalable, secure, explainable, and sustainable intelligent IoT ecosystems.