AI-Powered Dynamic Pricing Engine

A real-time pricing optimization tool that balances revenue maximization with customer fairness constraints.

Table of Contents

• Problem Statement
• Key Features
• Tech Stack
• Installation & Setup
• How to Run
• Project Architecture
• Future Roadmap
• License
• Contributors

Problem Statement

Retailers often struggle to price products effectively.

• Too High: Customers leave for competitors.
• Too Low: Revenue is left on the table.
• Volatile: Frequent price changes damage brand loyalty.

Our Solution: An AI engine that analyzes demand elasticity and competitor data to recommend the "Sweet Spot" price—maximizing profit while adhering to strict fairness guardrails (preventing price gouging).

Key Features

• Demand Forecasting: Uses XGBoost to predict sales volume based on price, weekends, and promotions.
• Fairness Constraints: Automatically caps price recommendations within ±20% of the market average to protect customer trust.
• Interactive Dashboard: A React-based control center to simulate market scenarios (e.g., "What if the competitor drops their price to $90?").
• Revenue Visualization: Real-time chart comparison of Static Pricing vs. AI Optimized Pricing.

Tech Stack

Backend (The Brain)

• Language: Python 3
• Framework: Flask (REST API)
• ML Engine: XGBoost, Scikit-Learn, Pandas

Frontend (The Face)

• Framework: React (Vite)
• UI Library: Material UI (MUI)
• Visualization: Chart.js

Installation & Setup

Prerequisites

• Node.js & npm installed
• Python 3.8+ installed

1. Clone the Repository

git clone https://github.com/joel-jain/pricing-engine-mvp.git
cd pricing-engine-mvp

2. Backend Setup

Navigate to the server folder, create a virtual environment, and install dependencies.

cd server

# Create Virtual Environment
python -m venv venv

# Activate Environment
# Windows:
venv\Scripts\activate
# Mac/Linux:
source venv/bin/activate

# Install Dependencies
pip install flask flask-cors pandas scikit-learn xgboost

3. Frontend Setup

Open a new terminal, navigate to the client folder, and install node modules.

cd ../client
npm install

How to Run

Step 1: Wake up the AI (Backend)

Before running the server, generate synthetic data and train the model.

# Ensure you are inside /server with venv activated
python generate_data.py   # Generates 1,000 rows of retail history
python train_model.py     # Trains the XGBoost model (look for "MAE" score)
python app.py             # Starts the API

Server will start at: http://127.0.0.1:5000

Step 2: Launch the Dashboard (Frontend)

In your second terminal (inside /client):

npm run dev

App will start at: http://localhost:5173

Project Architecture

pricing-engine-mvp/
├── client/                 # React Frontend
│   ├── src/
│   │   ├── components/     # Dashboard & UI Logic
│   │   └── assets/
├── server/                 # Flask Backend
│   ├���─ app.py              # Application Entry Point
│   ├── routes.py           # API Endpoints (/predict, /optimize)
│   ├── services.py         # Business Logic & Inference Engine
│   ├── train_model.py      # ML Training Script
│   └── generate_data.py    # Data Simulation Utility
└── README.md

Future Roadmap

• Docker Support: Containerize the application for easy deployment.
• Real Database: Replace CSV with PostgreSQL for persistent storage.
• Multi-Product Support: Scale the model to handle thousands of SKUs simultaneously.
• Reinforcement Learning: Implement RL agents to learn long-term pricing strategies.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributors

• Joel George Jain - Full Stack Developer