Penerapan metode Single Exponential Smoothing untuk memprediksi permintaan labu darah

Abstract

The Indonesian Red Cross (PMI) Blood Transfusion Unit (UTD), which is tasked with meeting the need for blood bags in Malang City is experiencing several problems. One of them is that the supply of blood bags is less than the demand for blood. Based on administrative data, demand for blood gourds in PMI Malang City will increase in 2022 compared to 2021 and 2020. To overcome limited supplies, PMI needs to make estimates to anticipate a spike in demand for blood bags. This research aims to determine the future demand for blood bags, or what is called forecasting. One method of time series forecasting is Single Exponential Smoothing. The data used is the demand for blood bags in 2020-2022 at PMI Malang City. To measure the error level of the forecasting model, this research uses Mean Absolute Percentage Error (MAPE) and Mean Absolute Error (MAE). The best MAPE value of 7.75% was obtained at α=0.5. The MAPE value can be categorized as very good because it is less than 10%. And based on MAE measurements, the best value is also obtained at α=0.5.