An artificial neural network based mathematical model for a stochastic health care facility location problem

Area of Science:

• Operations Research
• Healthcare Systems Engineering
• Emergency Medicine

Background:

• Trauma care systems face challenges due to stochastic patient demand and transfer times.
• Optimizing the location of trauma centers and helicopter stations is crucial for efficient emergency response.

Purpose of the Study:

• To develop a model for optimizing trauma center and helicopter station locations.
• To minimize total cost, patient transfer time, and patient waiting time within trauma centers.

Main Methods:

• A stochastic mixed-integer linear mathematical model was formulated.
• An artificial neural network (ANN) was trained using simulation to estimate complex waiting time objectives.
• A hybrid multi-objective algorithm based on a non-dominated sorting water flow algorithm was employed.

Main Results:

• The proposed method effectively addresses the complex optimization problem of trauma care system location.
• Computational results demonstrate the efficacy of combining simulation, ANN, and optimization techniques.
• The approach successfully balances cost, transfer time, and waiting time objectives.

Conclusions:

• The integration of simulation, artificial neural networks, and optimization offers a powerful solution for enhancing trauma care system efficiency.
• Strategic placement of resources can significantly improve patient outcomes in emergency medical services.
• This research provides a robust framework for optimizing healthcare logistics under uncertainty.