The increasing complexity of modern computing environments has created significant challenges in resource management, process scheduling, and system optimization within advanced operating systems. Traditional operating system management approaches often rely on static policies and deterministic scheduling mechanisms, which struggle to adapt to dynamic workloads and heterogeneous resource demands. To address these limitations, this study proposes an Explainable AI-enabled framework for predictive resource allocation and intelligent process optimization in advanced operating systems. The proposed framework integrates deep learning, reinforcement learning, Bayesian decision modeling, multi-modal data fusion, and Explainable Artificial Intelligence (XAI) to enable adaptive, transparent, and uncertainty-aware decision-making. Deep learning models are employed for workload forecasting and resource demand prediction, while reinforcement learning supports intelligent process scheduling and adaptive optimization. Bayesian inference enhances uncertainty handling, and XAI mechanisms provide feature importance analysis, confidence estimation, and interpretable decision rationales. Experimental evaluation demonstrates significant improvements over conventional operating system management approaches. The proposed framework achieved a decision accuracy of 91.6%, reduced latency from 67.9 ms to 38.2 ms, improved fault tolerance from 55.1% to 85.3%, enhanced uncertainty handling from 42.7% to 88.1%, and increased explainability scores from 30.5% to 79.4%. These results confirm that the integration of predictive analytics, intelligent process optimization, and explainable AI can significantly improve resource utilization, system responsiveness, reliability, and transparency. The proposed framework provides a scalable and trustworthy solution for next-generation intelligent operating systems and autonomous computing environments.
Sultan et al. (Mon,) studied this question.