IIT Madras · Business Data Management
A comprehensive academic capstone project focusing on optimizing inventory management and operational efficiency for Jai Maa Jhandewali Store, a family-run grocery business in Pitampura, Delhi. Built with rigorous data analysis using Excel records, ABC analysis, correlation studies, and machine learning forecasting for sustainable growth.
Roll No: 22f2000848
IIT Madras : BS in Data Science
Business Data Management - Capstone Project
Academic Project · Jan 2025 Term
Created as part of IIT Madras BS in Data Science & Applications
Family-Run Grocery Store (B2C Model)
Located in VP Block, Pitampura, North West Delhi, this family-run grocery store is owned by Mr. Satyam Prakash. Started with an initial investment of ₹7 Lakhs, the store achieved profitability after 1.5 years and serves the local community with dairy products, FMCG items, and groceries.
Family-run grocery stores face unique challenges in managing inventory efficiently while maintaining profitability. Jai Maa Jhandewali Store, with its limited resources and manual operations, experiences difficulties in inventory optimization, space utilization, and business growth. These challenges impact operational efficiency, customer satisfaction, and the store's ability to compete with modern retail formats.
Difficulty in maintaining optimal stock levels, leading to overstocking or stockouts, especially for perishable items, and lack of automated tracking systems for efficient inventory management.
Limited physical storage space resulting in inefficient product organization, clutter, and suboptimal use of available retail space affecting customer experience and operations.
Reliance on family-run operations with no automated systems for sales/inventory tracking and absence of home delivery service to compete with quick-commerce platforms.
Primary data collected from handwritten sales records digitized into Excel format, covering 13 months of operational data from Jan 2024 to Jan 2025.
8,563 bills aggregated into 398 rows × 14 columns covering Dairy, FMCG, and Groceries
Dairy (Milk, Butter), FMCG (Chips, Biscuits, Soft Drinks), Groceries (Rice, Pulses, Spices)
Comprehensive data processing from handwritten records to digital format, ensuring data quality and consistency for analysis.
Multiple analytical approaches including statistical analysis, trend analysis, ABC categorization, and machine learning forecasting.
Identified sales patterns and seasonal behaviors over time to understand demand fluctuations and prepare inventory accordingly.
Analysis: Temporal sales patterns and seasonal behavior analysis
Monthly sales trends showing seasonal patterns and festival peaks
Applied Linear Regression models to forecast future sales demand based on historical data patterns and seasonal trends.
Model: Linear Regression for demand forecasting
Linear Regression model predictions vs actual sales data for Milk
Categorized products based on their sales contribution to prioritize inventory management and space allocation.
Method: Pareto analysis for inventory prioritization
Pareto chart showing product categorization based on sales contribution
Examined relationships between different SKUs to identify complementary products and cross-selling opportunities.
Analysis: Pearson correlation matrix for cross-selling opportunities
Heatmap showing correlation coefficients between different product categories
Sales show clear seasonal patterns with predictable peaks during festivals and specific months. Dairy and FMCG categories exhibit highest fluctuations.
ABC analysis reveals 70.59% of sales come from just 6 products. Milk alone contributes 25.85% of total revenue while Category C items underperform.
Manual operations and space limitations restrict scalability. Owner's academic commitments create operational gaps during critical business periods.
