Emergency Medical Services Demand Forecasting: Modern Machine Learning Approaches For Producing Short-Term Spatiotemporal Estimations

Search for this publication on Google Scholar
Martin, R. (2019). Emergency Medical Services Demand Forecasting: Modern Machine Learning Approaches For Producing Short-Term Spatiotemporal Estimations. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

205 views ◎
163 downloads ⇓


Abstract

Emergency medical services (EMS); commonly referred to as ambulance, paramedic or pre-hospital emergency services, are a critical component in the delivery of urgent medical care to communities. EMS agencies, the organizations responsible for providing out-of-hospital acute medical care to the population of a specific service area, are confronted with the evolving task of effectively allocating the ambulances and medical personnel required to provide sufficient geographic coverage while minimizing response times to high-priority call requests. To meet this challenge, EMS practitioners and researchers have investigated the effectiveness of using various forecasting techniques for predicting future call volumes and demand densities. In this study, a forecasting methodology is proposed for producing spatiotemporal call volume predictions at a degree of granularity in time and space that is practical and actionable. A series of daily, hourly, and spatially distributed hourly call volume predictions are generated using a multi-layer perceptron (MLP) artificial neural network model following feature selection using an ensemble-based decision tree model. For spatially distributed predictions, K-Means clustering is applied to produce heterogeneous spatial clusters based on call location and associated call volume densities. The predictive performance of the MLP model is benchmarked against both a selection of traditional time-series forecasting techniques and a common industry method. Results show that MLP models outperform time-series and industry forecasting methods, particularly at finer levels of spatial granularity where the need for more accurate call volumes forecasts is more essential.

Details
Author

Martin, Richard

Title

Emergency Medical Services Demand Forecasting: Modern Machine Learning Approaches For Producing Short-Term Spatiotemporal Estimations

Physical Description

1 online resource (104 pages) : PDF

Date

2019

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Emergency medical services (EMS); commonly referred to as ambulance, paramedic or pre-hospital emergency services, are a critical component in the delivery of urgent medical care to communities. EMS agencies, the organizations responsible for providing out-of-hospital acute medical care to the population of a specific service area, are confronted with the evolving task of effectively allocating the ambulances and medical personnel required to provide sufficient geographic coverage while minimizing response times to high-priority call requests. To meet this challenge, EMS practitioners and researchers have investigated the effectiveness of using various forecasting techniques for predicting future call volumes and demand densities. In this study, a forecasting methodology is proposed for producing spatiotemporal call volume predictions at a degree of granularity in time and space that is practical and actionable. A series of daily, hourly, and spatially distributed hourly call volume predictions are generated using a multi-layer perceptron (MLP) artificial neural network model following feature selection using an ensemble-based decision tree model. For spatially distributed predictions, K-Means clustering is applied to produce heterogeneous spatial clusters based on call location and associated call volume densities. The predictive performance of the MLP model is benchmarked against both a selection of traditional time-series forecasting techniques and a common industry method. Results show that MLP models outperform time-series and industry forecasting methods, particularly at finer levels of spatial granularity where the need for more accurate call volumes forecasts is more essential.

Genre

doctoral dissertations

Subjects--Topics

Operations research
Information technology

Degree

Ph.D.

Keywords

Deep Learning
Emergency Medical Services
Forecasting
Machine Learning
Spatial Clustering

Subject Area

Information Technology

Advisor(s)

Saydam, Cem

Committee Members

Shehab, Mohamed
Ge, Yaorong
Ozelkan, Ertunga

Degree Note

Thesis (Ph.D.)--University of North Carolina at Charlotte, 2019.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Martin_uncc_0694D_12258

Permalink

http://hdl.handle.net/20.500.13093/etd:2595