Managing low-acuity patients in an Emergency Department through simulation-based multiobjective optimization using a...

Area of Science:

• Healthcare Management
• Operations Research
• Artificial Intelligence

Background:

• Emergency Departments (EDs) face overcrowding, particularly with low-acuity patients.
• Fast-track systems are implemented to manage patient flow and reduce wait times.
• Optimizing resource allocation in minor injury units is crucial for ED efficiency.

Purpose of the Study:

• To develop an efficient optimization strategy for resource allocation in ED minor injury units.
• To minimize patient waiting times and ED operating costs simultaneously.
• To address multiobjective simulation-based optimization with expensive black-box functions.

Main Methods:

• Formulated the problem as a multiobjective simulation-based optimization.
• Proposed a metamodeling approach using artificial neural networks (ANNs) to approximate black-box functions.
• Employed ANNs to replace time-consuming simulations for objective function evaluation.

Main Results:

• The ANN-based metamodeling approach efficiently solved the multiobjective optimization problem.
• A set of Pareto optimal solutions was generated for decision-making.
• Computational experiments on a real case study demonstrated the approach's reliability and effectiveness.

Conclusions:

• The proposed metamodeling approach significantly improves upon standard derivative-free optimization methods for ED resource allocation.
• This strategy offers a reliable and effective way to optimize ED operations and reduce patient wait times.
• Decision-makers can utilize the Pareto optimal solutions to tailor ED resource management to specific needs.