The integration of Explainable Artificial Intelligence (XAI) with edge computing offers a powerful approach for transparent real-time decision-making in Internet of Things (IoT) ecosystems. However, deploying complex XAI models on resource-constrained edge devices remains a significant challenge. This study proposes a novel framework that optimizes XAI methods for edge environments by simplifying model architectures and utilizing techniques like model pruning and quantization. The framework also adapts explainability tools such as SHAP and LIME to enhance interpretability without compromising performance. Focusing on applications in smart healthcare and industrial IoT, this research demonstrates how transparent AI decisions improve safety, reliability, and user trust. Furthermore, the study investigates the role of XAI in enhancing IoT security by detecting and mitigating real-time anomalies. Evaluations based on metrics such as processing speed, energy efficiency, and interpretability showcase the practicality and effectiveness of the proposed approach. This work bridges the gap between explainability and computational efficiency, paving the way for deploying trustworthy AI systems in resource-limited edge computing environments.