A hybrid genetic algorithm-queuing multi-compartment model for optimizing inpatient bed occupancy and associated...

Area of Science:

• Operations Research
• Health Systems Management
• Computational Biology

Background:

• Modern hospitals face challenges in optimizing bed occupancy and controlling utilization costs.
• Intelligent decision support systems are needed for efficient hospital management.

Purpose of the Study:

• To explore how easily understandable and implementable intelligent decision support systems can help hospital managers optimize bed occupancy and utilization costs.
• To propose a hybrid genetic algorithm-queuing multi-compartment model for patient flow optimization.

Main Methods:

• A hybrid model combining a finite capacity queuing model with phase-type service distribution and a compartmental model.
• An evolutionary-based approach using a genetic algorithm to optimize bed management and costs.
• Utilized historical bed-occupancy data from a hospital's Department of Geriatric Medicine.

Main Results:

• The hybrid model demonstrated that high bed occupancy (over 91%) is achievable with optimized bed allocation, impacting costs significantly.
• Different bed allocations resulted in substantial cost variations (£755 vs. £1172).
• Finite capacity systems showed lower costs than traditional queuing models at higher patient arrival rates.

Conclusions:

• Genetic algorithms are effective tools for optimizing hospital bed allocation and associated costs by encoding queuing and cost model information.
• The proposed methodology is adaptable for various medical departments with minor adjustments.
• Intelligent decision support systems offer a pathway to improved hospital operational efficiency and cost control.