Pembelian Barang Setiap Bulan di Warung Delstore Menggunakan Metode Regresi Linier Berganda Berbasis Web

Abstract

The purpose of the study was to develop a monthly goods purchase prediction system for Warung Delstore using a web-based multiple linear regression method. Micro, Small, and Medium Enterprises (MSMEs) like Warung Delstore play a crucial role in the economy, yet their inventory management is often inefficient due to a lack of information technology implementation. Through this prediction system, shop owners can more accurately estimate stock needs, thereby reducing the risk of stock shortages that could disappoint customers or surplus inventory that remains unsold. The system development method employed was the prototype method, where the system was developed interactively by gathering user feedback at each stage. The stages in this method included communication with users to determine the required software, followed by creating a model in the second stage, which involved rapid prototyping. In the third stage, a program is developed using multiple linear regression as a tool to predict the prototype's outcomes, followed by testing. After testing, the program was handed over to users for trial and maintenance. The result of this research was a web-based prediction software that was utilized by Warung Delstore's owners to manage inventory more effectively. The system was equipped with a prediction feature capable of providing recommendations for the required monthly stock quantities based on previous sales data