PERBANDINGAN METODE EXPONENTIAL SMOOTHING DAN METODE DEKOMPOSISI UNTUK PERAMALAN JUMLAH PENGUNJUNG PERPUSTAKAAN DAERAH PROVINSI SUMATERA SELATAN
Daftar Isi:
- The library is a center of information that must be managed properly in order to provide maximum benefits. One application of forecasting method is predicting the number of library visitors. This study aims to obtain a comparison of exponential smoothing holt winters method and decomposition method in predicting the number of visitors of Regional Library of South Sumatera Province, and predicting the number of visitors of Regional Library of South Sumatera Province from January to December (Period 61-72) Year 2017. To achieve the objective the research used exponential smoothing holt winters method and decomposition method by doing error calculation consisting of mean absolute error (MAPE), mean absolute deviation (MAD), and mean square deviation (MSD). Based on the results of data analysis, the best forecasting method is known to predict the number of visitors of Povinsi Regional Library of South Sumatra is exponential smoothing method holt winters. This method is said to be better because it has a smaller forecast error value when compared with the value of forecast error Decomposition method. Exponential Smoothing Holt Winters method uses smoothing constants α = 0,1, γ = 0,1, and β = 0,1 yields MAPE = 16, MAD = 1338, and MSD = 6994313.