E-CommerceMongoDB

This project implements a NoSQL document-oriented database in MongoDB for a UK-based online shopping and grocery delivery platform. It was developed to simulate the backend data architecture and functionality for an e-commerce business expanding into fresh grocery delivery via a partner model.

The repository includes MongoDB collection modeling, realistic operational queries, data seeding scripts, and command-line interfaces to simulate backend operations like order placement, partner assignment, inventory tracking, and personalized recommendations.

Background

The platform merges two business models:

1. Regular Product Orders – Standard e-commerce orders for books, CDs, phones, and appliances.
2. Fresh Product Orders – Same-day or instant delivery groceries from local stores in Manchester, handled by delivery partners.

The database is designed to support:

• Multiple product categories with complex metadata
• Fresh product logistics using geo-coordinates
• Real-time order processing and driver assignment
• Customer recommendations based on ratings
• Inventory management

Setup Instructions

Activate environment

conda activate mongo_venv

MongoDB Atlas Setup

• Create a MongoDB Atlas cluster (or run locally)
• Create a database named Amazone
• In scripts/db_connect.py, update your connection string:

conn_str = "mongodb+srv://<ID>:<password>@udatabases.a2oqj.mongodb.net/?retryWrites=true&w=majority&appName=UDatabases"

• Create collections by loading sample data from the data/ folder or through this script

python scripts/seed_database.py

Database Collections

Collection Name	Description
amazone_partners	Delivery partners/drivers with status and live location
books	Book products with metadata like author, publisher, ISBN
cds	CD products with artist, track count, and duration
current_orders	Active (pending/confirmed) customer orders
customer_address	One-to-many addresses linked to customers (billing/shipping)
customers	Customer data including demographics, ID, and recommendation list
daily_inventory_level	Historical inventory levels by warehouse and product
delivery_records	Records of past deliveries, possibly for partner payout tracking
fresh_products	Grocery items linked to stores
home_appliances	Appliance products with specs like voltage, style, and color
mobile_phones	Phone products with brand, model, color, and features
past_orders	Delivered orders including total cost and order items
products	Unified product master data used for lookups and categorization
ratings	Customer-product rating documents (sparse but frequently queried)
regular_products	Non-grocery products (books, CDs, phones, appliances)
stores	Store metadata with geo-coordinates and available inventory

Running the Queries

Each query is implemented as a standalone CLI script inside the queries/ folder.

Each script accepts arguments via --flags.

insert_fresh_product.py

Function: Add a fresh grocery product to a customer’s cart. If a pending order already exists, update it; else, create a new order.

Usage:

python -m queries.insert_fresh_product --customer_id 1 --store_id 1 --product_id 45 --quantity 1

confirm_order.py

Function: Confirm a fresh order and assign the nearest available delivery partner (based on store and driver location).

Usage:

python -m queries.confirm_order --customer_id 1 --store_id 1

get_order.py

Function: Fetch detailed info about a shipped fresh order, including products, prices, delivery partner info, and ETA.

Usage:

python -m queries.get_order --order_id 2025

list_fresh_products.py

Function: Retrieve fresh grocery products available in nearby stores based on user’s latitude and longitude.

Usage:

python -m queries.list_fresh_products --latitude 53.4668 --longitude -2.2339

sales_by_category.py

Function: Calculate total sales revenue grouped by product category (across fresh and regular products).

Usage:

python -m queries.sales_by_category

inventory_by_warehouse.py

Function: Aggregate and visualize total inventory quantity stored in each warehouse using a pie chart.

Usage:

python -m queries.inventory_by_warehouse

process_delivered_order.py

Function: Move a completed order from current_orders to past_orders, log delivery time, and update driver status.

Usage:

python -m queries.process_delivered_order --order_id 2025

customer_recommendations.py

Function: Fetch top recommended products for a customer based on predicted ratings.

Usage:

python -m queries.customer_recommendations --customer_id 123

product_rating.py

Function: Calculate the average rating and total number of ratings for each product.

Usage:

python -m queries.product_rating

top_customers.py

Function: Display the top 10 customers ranked by total spending.

Usage:

python -m queries.top_customers