An Integrated Intelligent Decision Support System Combining Stochastic Programming and Queuing Theory for Dynamic Triage of Surgical Critical Care Patients to Prevent Postoperative Complications in Healthcare Business Operation

Matching surgical patients to intensive care beds in real time remains stubbornly difficult in healthcare business operation. Demand spikes. Capacity tightens. Yet, even brief admission delays might trigger avoidable complications and inflate operational costs. To navigate this bottleneck, this study proposes the model of Stochastic Programming and Queueing Theory-Based Decision Support for Dynamic Surgical Triage (SQDSS), a framework blending two-stage stochastic programming with multi-class priority queueing. It attempts to capture uncertain arrivals and shifting acuity, alongside the grim reality that waiting patients often deteriorate. This investigation tested the model across 48 synthetic scenarios. Subsequently, the testing results exhibited that SQDSS experimentally appears to outperform both first-come, first-served and fixed-threshold policies. Empirically, high-acuity waits seemingly dropped by roughly 23%, while bed utilization potentially climbed by 16%. Queue overflows and simulated complication rates also suggested notable declines, hovering near 28% and 18%, respectively. Interestingly, the most pronounced gains emerged during surge conditions, particularly when trauma cases dominated. While no mathematical abstraction perfectly mirrors clinical reality, these findings hint that merging stochastic optimization with queueing analytics could offer a robust foundation for dynamic triage. Ultimately, such hybrid approaches might just foster more resilient healthcare business operation and supply chain management.