Author/Authors :
Asfora Frej, Eduarda Universidade Federal de Pernambuco - Av. Acadêmico Hélio Ramos - s/n-Cidade Universitária - Recife, Brazil , Peixoto Roselli, Lucia Reis Universidade Federal de Pernambuco - Av. Acadêmico Hélio Ramos - s/n-Cidade Universitária - Recife, Brazil , Pires Ferreira, Rodrigo José Universidade Federal de Pernambuco - Av. Acadêmico Hélio Ramos - s/n-Cidade Universitária - Recife, Brazil , Ramalho Alberti, Alexandre Universidade Federal de Pernambuco - Av. Acadêmico Hélio Ramos - s/n-Cidade Universitária - Recife, Brazil , Teixeira de Almeida, Adiel Universidade Federal de Pernambuco - Av. Acadêmico Hélio Ramos - s/n-Cidade Universitária - Recife, Brazil
Abstract :
This paper puts forward a decision model for allocation of intensive care unit (ICU) beds under scarce resources in healthcare
systems during the COVID-19 pandemic. The model is built upon a portfolio selection approach under the concepts of the
Utility Theory. A binary integer optimization model is developed in order to find the best allocation for ICU beds, considering
candidate patients with suspected/confirmed COVID-19. Experts’ subjective knowledge and prior probabilities are considered to
estimate the input data for the proposed model, considering the particular aspects of the decision problem. Since the chances of
survival of patients in several scenarios may not be precisely defined due to the inherent subjectivity of such kinds of
information, the proposed model works based on imprecise information provided by users. A Monte-Carlo simulation is
performed to build a recommendation, and a robustness index is computed for each alternative according to its performance as
evidenced by the results of the simulation.
