Determination of an optimal forecast model for ambulance demand using goal programming

Area of Science:

• Operations Research
• Public Health Management
• Forecasting Science

Background:

• Accurate forecasting of emergency medical service (EMS) demand is vital for efficient resource allocation.
• Existing forecasting models may not adequately prioritize critical emergency calls over routine ones.

Purpose of the Study:

• To develop and evaluate a multistep approach for optimizing exponential smoothing model parameters to forecast EMS demand.
• To improve the accuracy and cost-effectiveness of EMS demand forecasting by incorporating multiple objectives.

Main Methods:

• Utilized Winters' exponential smoothing model for daily emergency and routine demand data across four South Carolina counties.
• Formulated a goal programming model to prioritize accurate emergency call forecasting within the overall demand prediction.
• Optimized model parameters by minimizing the objective function value of the goal programming problem.

Main Results:

• The multistep, multi-objective approach yielded more accurate EMS demand forecasts compared to single-objective exponential smoothing and multiple linear regression models.
• The developed model implicitly weights demand by severity, providing a reliable overall demand estimate.
• The enhanced forecasting method proved more cost-effective for planning purposes.

Conclusions:

• The presented multi-objective goal programming approach offers a theoretical advancement in improving ambulance demand forecast accuracy.
• This method has the potential to significantly enhance EMS planning efficiency and resource management.
• The model demonstrates a practical application of operations research principles to public health service delivery.