A Dynamic User-Centric QFD-MCDM Model for Intelligent Cloud Service Selection

Cloud service selection is a complex multi-criteria decision-making process that must consider diverse user requirements such as cost, performance, availability, and scalability. Traditional approaches rely on fixed or static weight assignments for customer requirements (CRs), which limits their adaptability across different user contexts and application scenarios. To address this limitation, we propose a dynamic weight-driven framework that integrates Quality Function Deployment (QFD) with a multi-criteria decision-making (MCDM) method to enhance cloud service selection. In this framework, user-defined or task-specific preferences are dynamically converted into normalized CR weights. These weights are then used in a CR–TR (technical response) matrix to compute the relative importance of technical features using QFD. The resulting TR weights are applied within an MCDM model (e.g. TOPSIS) to rank cloud service providers based on both technical performance and contextual priorities. We validate the proposed approach using a real-world data set comprising ten commercial cloud service providers and a range of QoS metrics, including CPU performance, memory, disk I/O, and price. Experimental results demonstrate that the dynamic weighting mechanism significantly improves flexibility, personalization, and task awareness of the decision-making process, providing more accurate and user-aligned cloud service recommendations.