International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 03 | Mar 2025
p-ISSN: 2395-0072
www.irjet.net
iDairy Application : Product Demand Forecasting using SARIMA Model Ridham Jasani1, Jasmit Bajaria2, Vedant Deshmukh3, Dolas Keche4 1,2,3,4 Student, Computer Engineering, Rajiv Gandhi Institute of Technology, Maharashtra, India
---------------------------------------------------------------------***--------------------------------------------------------------------1.1 PROPOSED SYSTEM
Abstract - iDairy is a system designed to optimize dairy
retail operations through accurate demand forecasting. It utilizes advanced algorithms to analyze historical sales data and market trends, enabling precise inventory management and reducing wastage. The system is integrated with Google Sheets for real-time data storage, with each product having its own sub-sheet for efficient tracking. A Seasonal Autoregressive Integrated Moving Average (SARIMA) model is employed for forecasting, considering seasonal variations in demand. iDairy provides retailers with actionable insights to streamline stock replenishment and enhance operational efficiency. The project focuses on data-driven decision-making in dairy supply management.
The iDairy system is designed to improve demand forecasting and inventory management in dairy retail operations. The system utilizes historical sales data to predict future demand, helping retailers optimize stock levels and reduce wastage. The Seasonal Autoregressive Integrated Moving Average (SARIMA) model is used for forecasting, as it captures seasonal trends and variations in sales. Key features of the proposed system include:
Key Words: Machine Learning, Demand Forecasting, Inventory Management, SARIMA Model, Dairy Retail, DataDriven Decision Making, Sales Analysis.
Data Collection & Storage: Historical sales data is recorded in Google Sheets for easy tracking and analysis.
Demand Forecasting Model: SARIMA-based model predicts future sales based on past trends and seasonal variations.
Inventory Optimization: Forecasted demand helps maintain balanced stock levels, reducing the risk of overstocking or shortages.
1.INTRODUCTION The dairy industry faces significant challenges in managing inventory efficiently due to fluctuating demand patterns, seasonal variations, and perishable product constraints. Traditional inventory management methods often lead to overstocking or shortages, resulting in financial losses and operational inefficiencies. To address these issues, iDairy is developed as an AI-powered system that leverages Machine Learning (ML) for accurate demand forecasting and optimized inventory management.
By implementing iDairy, retailers can streamline supply chain management, minimize product wastage, and ensure efficient operations, ultimately improving profitability and resource utilization.
2. LITERATURE SURVEY
iDairy integrates a Seasonal Autoregressive Integrated Moving Average (SARIMA) model to analyse historical sales data, identify trends, and predict future demand with high precision. This approach helps retailers make data-driven decisions, reducing wastage while ensuring optimal stock levels. The system is designed to work seamlessly with Google Sheets, where each product has a dedicated sub-sheet for real-time data storage and tracking. This structured format enhances accessibility and allows businesses to monitor sales patterns efficiently.
Demand forecasting has been widely studied using both traditional statistical methods and deep learning techniques. ARIMA is effective for forecasting in stable environments with low variability, providing reliable predictions for stationary time-series data. However, it struggles with highly seasonal or complex datasets. On the other hand, LSTM models excel at capturing non-linear relationships, trends, and seasonality, making them ideal for more complex demand forecasting tasks. Studies show that training models on monthly data rather than weekly improves accuracy. Future research may focus on integrating attention mechanisms and global models to further enhance forecasting capabilities and optimize inventory management, particularly in industries like dairy production[1].
By implementing AI-driven analytics, iDairy aims to streamline supply chain processes, improve profitability, and support sustainable dairy retail operations. The project focuses on transforming traditional dairy management through automation, ensuring retailers can maintain a balanced inventory, minimize losses, and enhance overall operational efficiency.
© 2025, IRJET
|
Impact Factor value: 8.315
Demand forecasting using Business Intelligence (BI) and machine learning has gained significant attention in recent years. Various methods like time series analysis and rulebased forecasting models, such as Deep AR, have been explored for accurate predictions. Research shows that
|
ISO 9001:2008 Certified Journal
|
Page 31