Implementation Of Data Mining In Forecasting Herbal Medicine Sales At Cv. Anugerah Alam Indonesia Using The Linear Regression Method

Abstract

CV. Anugerah Alam Indonesia is a company engaged in the production and sale of herbal medicines. The main issue faced by the company is the inaccuracy in estimating production quantities, which often leads to either overstocking or stock shortages. This research aims to develop a sales forecasting application for herbal products using the simple linear regression method as part of data mining techniques. The sales data used in this study spans from 2020 to 2025 and focuses on ten selected herbal products. The application is web-based, developed using the PHP programming language and Laravel framework, with MySQL as the database. The forecasting process involves calculating regression constants and coefficients based on historical sales data and evaluating the prediction results using the Mean Absolute Percentage Error (MAPE) metric. The test results show that the forecasting model demonstrates varying levels of accuracy, ranging from very good to acceptable, depending on the product. By implementing this system, the company can optimize its production and distribution processes, avoid resource waste, and improve operational efficiency.