The increasing complexity of Amazon’s fulfillment center operations—driven by dynamic customer demand, extensive product variety, and high service-level expectations—necessitates the integration of real-time analytics and decision intelligence systems. This review explores the development of a comprehensive model that unifies predictive, prescriptive, and adaptive analytics to optimize core operational processes such as order picking, inventory allocation, workforce management, and last-mile delivery scheduling. By leveraging streaming data from IoT sensors, robotics, and warehouse management systems, real-time analytics enables continuous visibility into operational performance, while decision intelligence frameworks combine machine learning and simulation techniques to recommend optimal actions. The study synthesizes existing literature and case studies on AI-driven logistics, queue theory optimization, and cloud-based analytics infrastructures within large-scale e-commerce environments. It highlights key challenges such as data latency, model interpretability, and the alignment of algorithmic decisions with human oversight. Furthermore, the paper discusses the strategic implications of deploying real-time decision intelligence models for enhancing agility, sustainability, and customer satisfaction. The review concludes by proposing a hybrid framework integrating digital twins, reinforcement learning, and business intelligence dashboards to transform Amazon’s fulfillment centers into self-optimizing, adaptive systems capable of proactive operational governance.