Modelos Box & Jenkins aplicados a previsão de demanda de leitos hospitalares

Abstract

Since its foundation in 1970 the University Hospital of Santa Maria – HUSM is a reference in public health for the central region of Rio Grande do Sul. Being part of the Federal University of Santa Maria, the institution operates as a school hospital aiming at the teaching development research and health assistance. The main purpose of this research is to forecast the hospital occupation rate in sectors named: General Hospital, Emergency Room of HUSM, using the Box & Jenkins methodology and Intervention analysis, that will provide to future known of these variables at shortterm. The data were collected at the HUSM statistical sector, with monthly observations from January of 2000 to December of 2004. The model that better explain the occupation rate in Emergency Room was an autoregressive integrated moving average model with one difference, ARIMA(1,1,1) with one intervention of lag 36 of type abrupt temporary. Which give more explanation about the variable analyzed. To the General Hospital a autoregressive integrated moving average model with one difference, the variable was explained by ARIMA(1,1,2). Obtained the forecasted values will serve as support for future researches in this area, as well as the reached foresees will mainly become a way to evaluate the demand for beds at HUSM in the following months serving as basis for a better hospital organization.