APLIKASI PREDIKSI PENJUALAN UNTUK MENENTUKAN PERSEDIAAN STOK MENGGUNAKAN METODE LEAST SQUARE DI PT.BINTANG INDO JAYA

PT. Bintang Indo Jaya is a company engaged in the production of bakery (Rotiboy), for its own marketing Rotiboy distributed to several locations in the area of Surabaya and Bali. Rotiboy distribution is usually based on a request from each outlet Rotiboy are entrusted to employees working at the booth. Total demand for highly precise effect on the smooth sales process Rotiboy. Hence the absence of computerized systems for decision support in determining demand for inventory stock of bread, outlets often lack and excess stock. This research applies data mining techniques prediction by using the least squares method to predict the number of sales in the next month. Attributes are used are sales and a period of time. The data used are taken from data from PT. Bintang Indo Jaya Branch Gresik in 2014 and 2015 as many as 24 data. System testing is done by performing 10 tests using the data used to predict as many as 2 months, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months and 11 months. Then from the 10 test results were compared to obtain Forecast Error smallest to earn high accuracy. At the highest accuracy of the test results obtained on testing using the data four months amounted to 89.9% with the prediction error rate of 10.1%