Ride Hailing Simulation:

A Python ride-hailing simulation that combines JSON data ingestion, validation, SQLite persistence, graph-based driver matching, payment simulation, and analytics.

Overview:

This project simulates a ride-hailing system similar to Uber or Lyft. It demonstrates how structured data can be loaded from JSON, validated, persisted in a relational database, and used to perform graph-based driver matching and payment processing.

The system models real-world components including riders, drivers, rides, matching logic, and payments, providing a complete end-to-end data pipeline.

Tech Stack:

• Python
• SQLite
• NetworkX (graph-based routing)
• JSON (data ingestion)

Features:

• Load riders, drivers, and rides from JSON files
• Validate incoming data before use
• Persist riders, drivers, rides, and matching logs in SQLite
• Match riders to drivers using graph-based shortest path algorithms:
  • Unweighted shortest path (fewest hops)
  • Weighted shortest path (lowest cost)
• Simulate ride creation and payment processing
• Generate a simple analytics summary

The matching system uses graph algorithms to simulate real-world routing, enabling both efficiency-based and cost-based driver selection.

Project Structure:

simulation.py # Main entry point database_manager.py # SQLite connection and query helper repositories.py # Persistence functions for domain objects data_loader.py # JSON loading and ingestion validators.py # JSON validation helpers graph_model.py # Graph creation and edge weighting matching.py # Driver matching algorithms

rider.py driver.py ride.py car.py payment.py billing_info.py # Domain logic

schema.sql # Database schema data/ # Seed JSON files

Design Highlights:

• Modular architecture (data loading, validation, persistence, matching)
• Separation of concerns using a repository pattern
• Input validation layer to prevent malformed or inconsistent data
• Extensible design for adding new features (payments, analytics, preferences)

Example Output:

Database and tables created successfully.
Loaded 10 drivers into DB
Loaded 10 riders into DB
Loaded 10 rides into DB

--- Real matching test ---

UNWEIGHTED MATCH
Driver: Bob
Hops: 4
Path: [12, 72, 7, 92, 5]

WEIGHTED MATCH
Driver: Bob
Cost: 17
Path: [12, 72, 7, 92, 5]

Saved matching log ID: 1
Saved matched ride ID: 11
Loaded matched ride: Aisha has requested a ride from 12 to 25.

--- Payment Simulation ---
Charging rider Aisha $42.5...
Charging $42.50 to pm_1 for Aisha.
Payment successful ✅
Charge result: True

--- Analytics Summary ---
Total rides: 11
Completed rides: 10
Average fare: $24.73
Most active driver: Bob (1 rides)

How to Run:

1. Make sure Python 3 is installed
2. Install dependencies:
   • pip install networkx
3. Run the simulation:
   • rm rides.db
   • python3 simulation.py

Future Improvements:

• Preference-aware matching (language, distance, ratings)
• Store payment transactions in the database
• Simulate concurrent ride requests
• Expand analytics (driver utilization, ride distribution, etc.)

Final Note:

This project demonstrates a complete data-driven workflow: JSON → validation → objects → database → matching → payment → analytics