Determination of server location in emergency care systems: an index proposal using data envelopment analysis and the hypercube queuing model

Enzo Barberio Mariano; Regiane Máximo Siqueira; Caio Vitor Beojone; Octaviano Rojas Luiz; João Victor Rojas Luiz; Hugo Henrique Dos Santos

doi:10.7717/peerj-cs.1637

Determination of server location in emergency care systems: an index proposal using data envelopment analysis and the hypercube queuing model

PeerJ Comput Sci. 2023 Nov 3:9:e1637. doi: 10.7717/peerj-cs.1637. eCollection 2023.

Authors

Enzo Barberio Mariano¹, Regiane Máximo Siqueira¹, Caio Vitor Beojone², Octaviano Rojas Luiz^{1

3}, João Victor Rojas Luiz¹, Hugo Henrique Dos Santos¹

Affiliations

¹ Department of Production Engineering, College of Engineering, São Paulo State University (UNESP), Bauru, São Paulo, Brazil.
² Urban Transport Systems Laboratory (LUTS), École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland.
³ Department of Engineering, Institute of Sciences and Engineering, São Paulo State University (Unesp), Itapeva, São Paulo, Brazil.

Abstract

The objective of this article is to propose a new composite index (CI) that helps to determine the most effective location of servers in an Emergency Care System (ECS), using Benefit of the Doubt (BoD)/Data Envelopment Analysis (DEA) and the Hypercube queuing model. The CI proposed was developed in four stages: (1) definition of a number of possible ECS configurations through the application of mathematical partitions and permutations; (2) application of the hypercube queuing model to determine performance parameters for each ECS configuration; (3) application of DEA/BoD to build the CI and generate performance rankings, and (4) evaluation of the rankings obtained to define the best configuration for the ECS analyzed. Data from two real cases from Brazil were used to assess the CI proposal. The results obtained confirm that: (a) the hypercube model could, relatively quickly, determine the configuration parameters generated; (b) the application of an appropriate DEA/BoD model enabled the different configurations to be ranked with good discrimination; (c) a pattern in the relationship between ambulance concentration and configuration effectiveness could be identified; and (d) the CI proposed would benefit ECS managers who are making resource location decisions.

Keywords: Benefit of the doubt; Data envelopment analysis; Emergency care systems; Hypercube queuing model; Location problems; Queuing theory.

Grants and funding

This work was supported by The São Paulo Research Foundation, FAPESP (No. 14/19051-7). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.