🚗 FareCast – Smart Ride Price & Wait Time Prediction Engine

🚀 FareCast is an AI-powered system that predicts ride fares and waiting times across multiple platforms (Uber, Ola, Rapido) and vehicle types (Bike, Auto, Car), along with a future price forecast to help users book at the cheapest time.

---

🎯 Problem Solved

Ride prices fluctuate heavily due to:

• Traffic
• Time of day
• Demand (rush hours)
• Weather conditions

❌ Users don’t know when to book ❌ No platform shows future price trends

👉 FareCast solves this by predicting both current and future fares + wait times

---

🧠 What I Built

• 🔮 ML models to predict:

  • Ride fare (per platform + vehicle)
  • Waiting time (per platform + vehicle)

• 📈 2-hour price forecasting engine

• ⚡ FastAPI backend deployed on cloud

• 🌐 Frontend (Streamlit / Web UI)

• 🔗 Real-world API integration ready (Maps, Weather)

---

🧩 Key Features

• ✅ Platform-wise predictions (Uber / Ola / Rapido)
• ✅ Vehicle-wise predictions (Bike / Auto / Car)
• ✅ Wait time estimation
• ✅ Future forecast (every 15 minutes)
• ✅ Best time recommendation
• ✅ Savings calculation

---

📊 Sample Output

Current Predictions

Vehicle	Platform	Price (₹)	Wait (min)
Bike	Ola	79	1
Bike	Rapido	74	2
Bike	Uber	74	2
Auto	Ola	149	2
Car	Uber	190	2

---

Forecast Trend (Next 2 Hours)

Time	Bike (₹)	Auto (₹)	Car (₹)
18:30	74	149	181
19:00	76	144	183
20:00	71	140	197

📉 Insight:

Best time to book: 20:00 Potential savings: ₹3

---

📸 Screenshots (Proof of Work)

🔹 API Working (Swagger UI)

🔹 Prediction Output (JSON Response)

🔹 Forecast Visualization

🔹 Frontend UI

Add your FareCast UI screenshot here

🧠 ML Approach

• Random Forest Classifier

  • Predicts ride availability

• Random Forest Regressor

  • Predicts:

    • Price per KM
    • Waiting time

---

📊 Features Used

• Time (Decimal Hour)
• Day / Weekend
• Rush Hour flag
• Distance
• Traffic Level
• Surge Value
• Temperature & Humidity
• Pickup / Drop Zones

---

⚙️ Tech Stack

• Backend: FastAPI
• ML: Scikit-learn
• Data Processing: Pandas, NumPy
• Deployment: Render
• Frontend: Streamlit / Web UI
• APIs Ready: Google Maps, Weather APIs

---

🚀 Live API

[https://farecast-ml-api.onrender.com/farecast]

---

📈 Why This Project Stands Out

• Not just prediction → decision-making system

• Combines:

  • Time-series thinking
  • ML regression
  • Real-world ride logic

• Production-ready architecture

---

🔮 Future Improvements

• Real-time traffic integration
• Weather API integration
• User-specific personalization
• Deep learning models (LSTM for trends)

---

👨‍💻 Author

Prince Singh B.Tech CSE (Data Science) Aspiring Data Scientist 🚀