The Great Bangalore Hackathon

Supply Demand Management and Reducing Ride Denials

📌 Problem Statement (Provided by Namma Yatri)

Ride denials and increased passenger wait times during peak hours disrupt urban mobility. The goal is to optimize the supply-demand balance by:

• Predicting and addressing driver-side concerns leading to ride denials.
• Encouraging drivers to accept more trips through incentives and rewards.
• Reducing passenger wait times by efficiently connecting them to available drivers.
• Implementing real-time surge balancing for a fairer ride distribution.

💡 Our Solution

🚀 1. Demand Prediction Model

• Developed an XGBoost Regressor to predict taxi demand in Bangalore.
• Created a synthetic dataset simulating peak-hour trends due to lack of public data.
• Enabled drivers to anticipate high-demand areas for better supply allocation.

🎮 2. Gamification for Drivers

• Introduced daily streaks, missions, and challenges to boost engagement.
• Drivers earn rewards, coupons, and discounts upon achieving milestones.
• Proposed partnerships with fuel, insurance, and service companies for meaningful incentives.

📅 3. Pre-Booking Commitment System

• Requires both passengers and drivers to commit a small percentage of the fare before confirming a ride.
• If either party cancels, the other is compensated through an exponential deduction model.
• Helps reduce last-minute cancellations and ensures commitment.

🔄 4. Ride Sorting Based on Driver Preferences

• Instead of strict ride filtering, drivers can sort available rides based on:
  • Shortest wait time
  • Ride duration
  • Fare price
• This ensures flexibility for drivers while keeping passenger wait times low.

🛠 Tech Stack

Frontend

• Next.js, React.js

Backend

• Express.js, WebSockets

Machine Learning & Data Processing

• Pandas, Scikit-learn, NumPy
• XGBoost for demand prediction

APIs Used

• Google Maps API
• Google Distance Matrix API

👨‍💻 Contributors

• Vaibhav PR
• Chethohaar
• Dhruva D
• Vijesh Shetty