Real-time hospital capacity estimation by off-line simulation and metamodeling

In a disaster event, real-time estimates of hospital capacity in the disaster region are useful for patient/ambulance routing, resource planning and emergency operations management. We develop a generic hospital simulation model that is capable of representing a wide range of hospitals with various sizes and capabilities. Designed simulation experiments using representative hospital parameter values are completed off-line. These simulation results are then regressed to form a steady-state model. The transient state is modeled by a sum of exponential functions. Combining both transient and steady-state models, a parametric metamodel capable of estimating hospital capacity in real time is developed.