A GIS-Based Demand Forecast Using Machine Learning for Emergency Medical Services

• DOI: 10.1061/9780784413616.203
• Publication Date: 2014-06-17T19:33:24Z
• Authors: Albert Y. Chen; Tsung-Yu Lu
• Keywords: Ambulance service

The objective for pre-hospital Emergency Medical Service (EMS) is to reach to, pick up, and deliver patients efficiently. By increasing the operational efficiency, survival rate of major trauma could potentially be improved. In this research, authors applied Moving Average, Artificial Neural Network, Linear Regression, Support Vector Machine forecast emergency medical demand. results from these approaches, as a reference, used pre-allocation ambulances. training models well validation conducted with data collected New Taipei City EMS. order represent performance forecast, introduced Geographic Information System (GIS) manage visualize spatial distribution demand error. Moreover, built flexible model which organizes user given size area time step. With easy use acceptable prediction performance, research outcome has its potential current practice.