The complexity and criticality of hospital supply chain operations demand agile and intelligent risk management strategies. This review explores the development and application of real-time risk assessment dashboards powered by machine learning (ML) within hospital supply chain management systems. With increasing demand variability, supply disruptions, and compliance requirements, traditional reactive models are no longer sufficient for ensuring uninterrupted availability of medical supplies. Machine learning offers predictive capabilities to identify emerging risks, forecast supply-demand mismatches, detect anomalies, and support data-driven decision-making. By integrating ML algorithms into interactive dashboards, healthcare institutions can achieve real-time visibility across procurement, inventory, distribution, and vendor management processes. These dashboards serve as early warning systems, enabling stakeholders to proactively mitigate operational, financial, and clinical risks. This paper systematically reviews existing frameworks, models, and case studies, identifying current limitations, evaluating ML techniques used, and proposing a roadmap for scalable, resilient, and adaptive hospital supply chain infrastructures enhanced by real-time intelligence.