The rapid growth of Industry 4.0 and the rise of smart cities have generated massive, heterogeneous datasets across transportation, healthcare, energy, and public services. Harnessing these data streams for efficient and sustainable urban management requires artificial intelligence (AI) solutions that are both accurate and transparent. Traditional AI models often function as opaque black boxes, limiting their acceptance in high-stakes urban governance. To address this challenge, we propose an Explainable Artificial Intelligence (XAI) framework that integrates scalable data analytics, interpretable machine learning, and visualization modules for real-time decision support in smart city environments. Built on distributed big data infrastructures, the system processes high-velocity data streams and employs feature attribution, rule-based explanations, and model-agnostic interpretability techniques to generate human-understandable insights. The framework is evaluated using the Smart Cities Index dataset from Kaggle, encompassing multi-domain indicators covering infrastructure, mobility, environmental performance, and governance metrics, with experiments conducted across classification and ranking tasks to assess both predictive performance and explanation fidelity. Case studies in smart energy management, intelligent traffic control, and public safety demonstrate that the framework enhances prediction accuracy while improving interpretability, stakeholder trust, and decision transparency. By bridging the gap between high-performance AI and human interpretability, the proposed approach supports accountable, fair, and sustainable smart city operations aligned with Industry 4.0 principles.
Xu et al. (Sat,) studied this question.