SUPPLY SHORTAGE FORECAST USING SPATIAL DATA MINING FOR EMS
Albert Y. Chen1 and Chen-Hsi Su2
1) Assistant Professor, Civil Engineering Department, National Taiwan University, Taipei, Taiwan
2) Graduate Research Assistant, Civil Engineering Department, National Taiwan University, Taipei, Taiwan
Abstract: The demand for pre-hospital Emergency Medical Services (EMS) and the supply of hospital beds fluctuate with time. Overcrowding in the Emergency Room (ER) has led to prolongation of waiting time for the ambulances with patients waiting for a bed. This prevents ambulances from leaving the hospital, and the performance of the pre-hospital EMS is affected. To deliver patients to the hospital, which has empty beds, a flexible model is proposed. In this model, the EMS demand is modeled with user desired spatial and temporal sizes. By integrating the hospital bed information with the demand modeling provides information to decision makers for the adjustment of ambulance deployment and destination hospital selection. The Geographic Information System (GIS) is utilized for the visualization of the spatial distribution of EMS demand and supply. A case study of ambulance demand prediction in New Taipei City is presented, which considers 3 years of pre-hospital EMS data as input. Time series analysis, artificial neural network, and digital signal processing methods are used for the forecast of the pre-hospital emergency medical demand. With tolerable prediction error, the proposed approach shows its potential to be applied to the current practice.
Keywords: Emergency Medical Services, Demand Forecast, Hospital Bed, Digital Signal Processing, Time Series Analysis, Geographic Information System
Bibliographical Reference:
Albert Y. Chen, Chen-Hsi Su. “SUPPLY SHORTAGE FORECAST USING SPATIAL DATA MINING FOR EMS.” In Proceedings of International Conference on Civil and Building Engineering Informatics (ICCBEI 2015), 123. Tokyo, Japan, 2015.