🛢️ e-HOIST: Heavy Oil Intelligence & Simulation Twin

Smart India Hackathon 2026
Team: Craft Coders
Problem Statement: PS 120 - Digital Twin for Baghewala Field (Oil India Limited)

📌 The Challenge

Oil India's Baghewala Field produces heavy crude (17-19° API) with low mobility, requiring thermal enhanced oil recovery (Cyclic Steam Stimulation - CSS) and artificial lift (Sucker Rod Pumps - SRP). Currently, CSS heating cycles and SRP pumping are managed as isolated operations.

As the reservoir cools after steam injection, oil viscosity spikes. Because the surface pump speed is not dynamically adjusted to match this thickening oil, the system suffers from rod floating, severe impact loading, frequent pump unsetting, and massive energy waste.

🚀 Our Solution

e-HOIST is an AI-enabled Digital Twin that integrates reservoir behavior, wellbore conditions, and surface pumping systems into a single predictive platform. By bridging subsurface thermodynamics with surface mechanical operations in real-time, it eliminates the guesswork of isolated operations.

Key Innovations

• Joint Optimization Engine: Uses XGBoost to predict reservoir cooling and dynamically recommends Variable Frequency Drive (VFD) and Strokes Per Minute (SPM) reductions to match real-time oil viscosity.
• Predictive Risk Detection: Actively monitors load cell telemetry to predict rod floating and impact loading before mechanical failure occurs.
• Thermal-to-Mechanical Intelligence: Calculates the exact steam volume and soak time needed for the next CSS cycle, drastically reducing the Steam-Oil Ratio (SOR).
• Command Center Dashboard: A React-based UI providing operators with 3D wellbore overviews, live telemetry, interactive GIS mapping, and 1-click automation for field adjustments.

💻 Technical Architecture

• Frontend: React.js, Vite, Tailwind CSS, Chart.js, React-Leaflet, Lucide-React
• Backend: Python, FastAPI, Uvicorn
• AI / ML: XGBoost, Scikit-learn, Pandas

⚙️ How to Run Locally

1. Start the AI Backend (FastAPI)

cd backend
python -m venv venv
# Activate the virtual environment
.\venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Start the server
uvicorn main:app --reload
API runs on http://localhost:8000

2. Start the Frontend Dashboard (React/Vite)
Open a new terminal window:

Bash
cd frontend
# Install node modules
npm install
# Start the development server
npm run dev