A Sales Forecasting System Based on Linear Regression Model and Artificial Neural System

Abstract

With the advancement of globalization, the market competition among enterprises has become increasingly intense. To win a good market, an enterprise must understand and grasp the laws of the market economy and accordingly predict the future of the market. Only by ensuring correct estimates can companies develop a right business plan and ultimately capture the market. The aim of this 
research work is to investigate into the problem of not making correct sales forecast as seen in most businesses today. The system at study in this work is the supermarket. The research work took place in Obowo LGA of Imo State, Nigeria. During the study, it was discovered that the management of the supermarket makes sales forecast using conventional means. The manager of the supermarket makes sales predictions relying on past sales data written in accounting books. This involves lots of paper work and also is time consuming. The purpose of this project is to develop a computer based sales forecasting system (software) for a supermarket, because the computer is faster and more accurate in processing data than humans, and this will help to solve the problems encountered in the traditional ways of sales forecasting which involves lots of paper work and also takes a lot of time to process. To achieve this, I have adopted two methodologies, the first method is the Linear Regression Model were by data from previous sales events are collected and stored in a database which can then be used as references when making decisions for future sales events. The second method is the Artificial Neural System also known as Machine Learning were by the computer behaves like humans by being able to collect and store data from previous sales events and also predict future sales without any human input. Two programming languages were used in developing the software: PHP and MYSQL, with php as front-end and mysql as the backend database. The system would be able to perform the following tasks: supplier registration and management, product registration and management, user management, customer management, sales report, receipt printing and sales forecasting. The resultant software is a sales forecasting system with an interface that is user-friendly whose data processing is very fast and accurate, with the ability to automatically update itself based on recent sales events. It can be entered using a username and password. It is accessible either by an administrator or staff.