Abstract
• Comprehensive model optimises surgery planning for better patient and staff outcomes.
• Focuses on social sustainability while minimising wait times and cancellations.
• Integrates meta-heuristics for robust scheduling under uncertain conditions
• Study shows increased ICU beds enhance surgery scheduling, improving outcomes.
This paper proposes a comprehensive surgery planning and scheduling model that considers both upstream and downstream units under uncertain conditions, emphasising social sustainability in the surgical process. The model aims to improve the overall well-being of both patients and healthcare staff by achieving three key objectives: minimising overtime and idle time in operating rooms to increase resource utilisation, maximising the interval between consecutive patients to ensure a smoother workflow, reducing staff fatigue, and increasing the number of scheduled surgeries. This approach helps reduce the total number of patients on waiting lists and shortens their wait time for surgery, ultimately promoting better healthcare outcomes and staff welfare. Scenario-based robust optimisation is used to evaluate duration and arrival uncertainties. Two tailored meta-heuristics, a Multi-Objective Firefly Algorithm (MOFA) and a Non-dominated Sorting Genetic Algorithm (NSGA), are adopted. The exact results, obtained using GAMS, are compared with the results of the outlined meta-heuristic approaches. Benchmarking shows that the heuristics maintain solution quality while reducing CPU time by up to 90%. Finally, a sensitivity analysis was conducted to evaluate how the number of beds in the Intensive Care Unit (ICU) and wards influences the number of patients scheduled for surgery and resource utilisation in upstream and downstream units. While increasing ICU bed capacity improves the number of surgeries that can be scheduled, the same effect is not observed for ward beds. The analysis highlights ICU beds as the critical bottleneck, informing strategic capacity investment. A case study was conducted in a hospital; implementation reduced the cancellation rate by 35 %, demonstrating the model’s practical impact.