An Improved Simulation-Based Simulated Annealing Algorithm For Solving Dynamic Patient Admission Scheduling Problem

Abstract

Healthcare services are crucial for national well-being and managing diseases, but face rising demand and resource constraints, necessitating optimization studies to improve the treatment quality, with patient admission scheduling being a key challenge in dynamic and uncertain environments. Ineffective scheduling leads to long waiting times, resource overutilization, and decreased patient satisfaction. To tackle this problem, this paper proposes an improved Simulation-based Simulated Annealing algorithm (SSA) to address the Dynamic Patient Admission Scheduling under Uncertainty (DPASU) problem with the aim of maximizing treatment efficiency and patient comfort. The algorithm integrates a Simulation-Optimization approach with Simulated Annealing to adaptively assign hospital beds to patients while considering uncertainties in patient health priorities and lengths of stay. The performance of the proposed SSA approach has been tested on various datasets of the benchmark instances. Comparative analyses demonstrate that SSA outperforms traditional rule-based simulation methods, indicating its capability to offer near-optimal solutions and emphasizing the algorithm's applicability across different dataset sizes. Therefore, the improved SSA is proved to be an effective method to solve dynamic scheduling problems in the healthcare sector