Machine Learning-Based Sales Forecasting: Optimizing Product Demand Prediction

Abstract

Machine learning [ML] has become a transformative tool in business analytics, enabling firms to extract valuable insights, optimise oper ations, and enhance decision-making. Businesses widely use ML for supply chain management, fraud detection, customer segmentation, and sales forecasting. However, challenges such as poor data qual ity, high processing costs, integration constraints, and the need for skilled professionals often hinder its implementation. Addressing these challenges is crucial to unlocking the full potential of ML and driving efficiency, profitability, and innovation. This study applies ML techniques to forecast sales of jackets and blazers at DGN Clothing, Ludhiana, a major apparel manufacturer. By analysing historical sales data, seasonal trends, and pricing variations, this research explores how ML can help businesses optimise inventory, reduce losses, and improve profitability. The study evaluates Random Forest Regression to determine the most effective model for predict ing sales trends. Performance metrics such as Mean Absolute Error (MAE), R-Squared (R²), and Root Mean Squared Error (RMSE) are used to assess accuracy. The results offer actionable insights into demand fluctuations, pricing strategies, and consumer purchasing behavior, allowing businesses to make data-driven decisions. By leveraging MLbased sales forecasting, retailers can enhance stock management, minimise overproduction, and boost revenue generation. This research underscores the impor tance of ML in modern business analytics, demonstrating its ability to streamline processes and provide a competitive edge in the evolving apparel industry.

DOI: https://doi.org/10.24321/2456.429X.202503