Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{194902,
        author = {Sathya R and Yazhini Mathankumar},
        title = {A Supervised Machine Learning Framework for Big Mart Retail Demand Forecasting and Price Sensitivity Modelling},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {5654-5662},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=194902},
        abstract = {Predictive analysis for Big Mart sales using machine learning focuses on forecasting product-level sales by leveraging historical retail data and advanced computational techniques implemented in Python. The objective of this study is to assist retailers in making data-driven decisions related to inventory management, pricing strategies, and store performance optimization. By integrating machine learning models into a web-based system, the solution enables real-time sales prediction based on various product and outlet attributes. The proposed system utilizes a supervised machine learning regression model trained on structured retail data that includes item characteristics such as weight, fat content, visibility, product category, and maximum retail price (MRP). In addition, outlet-related features like store size, location type, outlet type, and establishment year are incorporated to capture the impact of store dynamics on sales performance.
 These features collectively help in modeling complex relationships between products and their selling environments. Python serves as the core development language due to its extensive ecosystem for data analysis and machine learning. Libraries such as NumPy and Pandas are used for data preprocessing and feature handling, while Joblib is employed to load the trained model efficiently for inference. The trained model is seamlessly integrated into a Django-based web application, allowing predictions to be generated dynamically from user-provided inputs. The system includes robust input validation and preprocessing mechanisms to ensure data consistency and reliability before feeding inputs into the predictive model. Numeric parsing, categorical encoding, and year validation are performed to minimize prediction errors and enhance model stability. This structured pipeline ensures that the predictions remain accurate and meaningful in real-world usage scenarios. Beyond generating a single sales prediction, the application extends its analytical capability by performing scenario-based analysis. By varying the MRP within a realistic range, the system produces multiple predicted sales values, enabling users to visualize how price fluctuations may influence demand. This predictive visualization supports better pricing and marketing decisions. To enhance interpretability and user engagement, the predicted results are presented using interactive visualizations such as line charts and tabular summaries.
 Chart-based representations of MRP versus predicted sales help users identify trends and patterns intuitively. These visual tools transform raw numerical outputs into actionable business insights. Overall, the predictive analysis framework for Big Mart sales demonstrates how machine learning, when combined with Python-based web technologies, can deliver practical and scalable retail intelligence solutions. The system highlights the effectiveness of predictive modeling in understanding consumer behaviour and optimizing retail operations, ultimately contributing to improved profitability and strategic planning in the retail sector.},
        keywords = {Predictive Analytics, Big Mart Sales Forecasting, Machine Learning Regression, Django Web Application, Retail Data Analysis, Scenario-Based Pricing Analysis},
        month = {March},
        }