<![CDATA[Hard drive failures remain a critical reliability concern in large-scale cloud data centres because they can lead to data loss, service downtime, and increased operational costs. Traditional threshold-based monitoring techniques often fail to capture nonlinear relationships among hard drive health indicators and may produce high false-positive rates. This study presents a conceptual framework for developing an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based hard drive failure prediction model using selected Self-Monitoring, Analysis, and Reporting Technology (SMART) attributes. It further examines the potential impact of key SMART indicators on predictive performance. By integrating fuzzy logic reasoning with neural network learning, the proposed framework is designed to improve predictive accuracy while maintaining interpretability. The study concludes that an ANFIS-based prediction framework can support proactive maintenance strategies for cloud service providers by enabling earlier identification of potential hard drive failures. This framework contributes to the development of intelligent predictive maintenance systems in cloud computing environments and offers practical implications for improving system reliability, reducing downtime, and enhancing operational efficiency.]]>
