Exponential Smoothing Model Using Statistical Software to Forecast Peritoneal Dialysis Sessions

In Peru, one of the most relevant problems in the field of health is the shortage of medical resources; based on a more particular and delicate approach, the resources related to peritoneal dialysis sessions are being affected day by day. The exponential smoothing model has been a very useful tool for the forecast of care and procedures to be performed in different health areas; in the present research, this tool was applied to determine the forecast of peritoneal dialysis sessions in a representative health entity in Peru. The results presented accuracy measures MAPE = 30.2, MAD = 116.6 and MSD = 392,473.7, which represent acceptable values according to the volume of historical information collected; also, the experimental error in relation to the forecast for week 48 was 3.54%. There were outliers which increased the value of the accuracy measures, and it is advisable to analyze the causes to optimize the model; however, the residual analysis presented favorable values in relation to the model fit. The applied forecasting model presented very favorable general results, which, under the support of Minitab software, did not present any difficulty in execution.