A Genetic Algorithm-Based Scheduling Framework for Hospital Resource Allocation in Edge-Terminal Collaborative Networks Qinglin Shi

Under the background of increasing medical demand and digital transformation, hospital resource scheduling in edge-terminal collaborative environments faces challenges such as information delay and insufficient decision-making accuracy. This study proposes a hospital resource allocation and scheduling framework based on genetic algorithm, which integrates edge terminal collaboration architecture, and heuristic optimization logic. The edge computing node is deployed in key medical areas, with a 1000 Mbps transmission rate and a 5 ms low latency hybrid network, to achieve real-time data acquisition and preprocessing of the terminal. The core of genetic algorithm adopts resource processing capability chromosome encoding, with a crossover probability of 0.7 and a mutation probability of 0.1. It minimizes task completion time, patient waiting time, and resource conflict cost through multi-objective fitness functions (α=0.4, β=0.3, γ=0.3). The validation results based on the MIMIC-III (Medical Information Mart for Intensive Care III) dataset show that the model's basic performance (accuracy 87%, F1 value 86%) and resource management effectiveness (accuracy 89%, F1 value 88%) outperform traditional FCFS (First Come, First Served) scheduling models, with an overall performance improvement of 2% -14%. This framework effectively improves resource utilization and service efficiency, provides practical solutions for optimizing hospital resource scheduling, and expands the application scenarios of heuristic algorithms in the medical field.