Implementasi Metode Support Vector Regression (SVR) Dalam Peramalan Penjualan Roti (Studi Kasus: Harum Bakery)

Noval Dini Maulana (Fakultas Ilmu Komputer, Universitas Brawijaya)
Budi Darma Setiawan (Fakultas Ilmu Komputer, Universitas Brawijaya)
Candra Dewi (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 3 No 3 (2019): Maret 2019

Bread is one of the favorite foods of the Indonesian people, the proof is the increasing import of wheat flour. One of the bakery companies that is currently developing is Harum Bakery. Constraints that are often faced by Harum Bakery are customer demand forecasting systems that are still manual and seem to be guessing. The forecasting process give a big impact on the sales process. With the forecasting of bread sales, it is hoped that Harum Bakery can help bakeries in preparing raw materials and everything needed for bread making. Support Vector Regression (SVR) is one method that can be used in forecasting. The data used is data on sales of sweet bread, cake and white bread with time series data types and uses 4 features. In this study the SVR method used to predict the results of the sale resulted in an evaluation value of RMSE for sweet breads is 0.00176, bread cake is 0.00019, and large breads is 0.00010.