Evaluating Bio-Inspired Metaheuristics for Dynamic Surgical Scheduling: A Resilient Three-Stage Flow Shop Model Under...

Area of Science:

• Operations Research
• Computer Science
• Healthcare Management

Background:

• Surgical scheduling requires balancing elective procedures with unpredictable emergency arrivals.
• Existing optimization methods may not adequately address dynamic scheduling needs and emergency integration.
• Improving operating room efficiency is crucial for reducing surgical backlogs.

Purpose of the Study:

• To evaluate optimization algorithms for surgical scheduling in a dynamic flow shop environment.
• To compare the performance of Genetic Algorithm with discretized bio-inspired algorithms (PSO, SBOA, MSO).
• To assess the algorithms' effectiveness in integrating emergency surgical cases without disrupting elective schedules.

Main Methods:

• Utilized a dynamic three-stage flexible flow shop model under no-buffer blocking constraints.
• Implemented and tested a Genetic Algorithm (GA) and discretized variants of Particle Swarm Optimization (PSO), Secretary Bird Optimization Algorithm (SBOA), and Mantis Shrimp Optimization Algorithm (MSOA).
• Conducted 300 Monte Carlo replications to analyze algorithm performance and efficiency.

Main Results:

• The Genetic Algorithm achieved the highest global efficiency.
• Discretized bio-inspired algorithms reached a comparable statistical efficiency frontier.
• The discretized Secretary Bird Optimization Algorithm demonstrated superior emergency integration by preserving capacity buffers, avoiding resource saturation.

Conclusions:

• Bio-inspired algorithms, especially SBOA, offer a viable alternative for optimizing surgical scheduling and improving emergency responsiveness.
• Optimized scheduling can lead to significant gains in operating room utilization, estimated at 90 hours annually.
• These improvements can provide crucial capacity to address non-prioritized surgical backlogs in public hospitals.