PENGGUNAAN METODE LINEAR REGRESSION UNTUK PREDIKSI PENJUALAN SMARTPHONE

Abstract

Planning and analyzing market needs precisely and efficiently if managed optimally is needed to achieve company success. In practice, existing transaction data used as a reference in planning and analyzing market needs. This company needs a tool to predict future sales. This information is needed because a good sales prediction will help understand what items must be distributed according to market needs so that companies can reduce uncertainty in decision making. The purpose of this study is to create an information system can do smartphone sales forecasting at 82 Cell Mayang with the Linear Regression method. The sales forecasting system is using the Linear Regression method, The goal to create a sales forecasting system by determining the sales volume with a certain period by looking at the cost of advertising and the number of sales. The research method used includes observation, interviews and literature studies. Designing using DAD includes entity relation diagrams, context diagrams, the hierarchy of input process output and data flow diagrams. The programming language used is Visual Basic.Net and the sql server 2008 database. Features in the sales forecasting application include processing data items, customer data, incoming product data, sales data, and forecasting data. The test results show the MAPE value is 0.032 and the MSE value is 5.16. From this value, it can be said that the prediction of smartphone sales with the Linear Regression method on 82 Cell Mayang is categorized as very good. Whereas for the blackbox testing that has been carried out, it shows that the smartphone sales forecasting system in 82 Cell Mayang, Sukoharjo has been going well.Keywords: forecasting, sales prediction, incoming product data, linear regression, visual basic