🤖 AI Inventory Management System

A full-stack AI-powered inventory management system with demand forecasting, real-time analytics, automated alerts, and role-based access control.

---

🌐 Live Demo

🔗 Frontend: https://your-app.vercel.app
🔗 Backend API: https://your-backend.render.com/docs

---

✨ Features

🔐 Authentication & Security

• JWT-based login with Remember Me (7 days)
• OTP-based forgot password via Gmail SMTP
• Admin notification on every password reset request
• Role-based access: Admin Manager Staff Viewer
• bcrypt password hashing

📦 Inventory Management

• Add, edit, soft-delete products
• SKU-based duplicate detection
• Bulk CSV upload with row-wise error reporting
• Stock history with full audit trail (before/after values)
• Reorder point & reorder quantity per product

📊 AI Demand Forecasting

• ARIMA model with auto order selection
• Per-product forecast with confidence intervals
• Days-of-stock-left prediction
• Fallback to Moving Average for short data series

📈 Analytics Dashboard

• KPI cards: inventory value, SKUs, orders, out-of-stock count
• Bar chart, Pie chart, Line chart (Recharts)
• Top selling products, turnover rate, stock status table
• AI-generated insights (danger / warning / info)

🔔 Smart Notifications

• Priority-based alerts: Critical → Warning → Info
• Out-of-stock, low stock, dead stock detection
• Lightweight navbar badge (polling every 30s)

🤖 AI Assistant

• Natural language inventory queries
• Rule-based engine: stock levels, value, trends
• Inventory snapshot sidebar
• Quick-ask prompts

👥 User Management

• Create / update / deactivate users
• Inline role change from table
• Last login tracking

---

🛠️ Tech Stack

Layer	Technology
Frontend	React.js, TailwindCSS, Recharts, Framer Motion
Backend	FastAPI, SQLAlchemy, SQLite
Auth	JWT (PyJWT), bcrypt
ML	ARIMA (statsmodels), pandas, numpy
Email	Gmail SMTP (smtplib)
Deploy	Vercel (frontend), Render (backend)

---

📁 Project Structure

ai-inventory-system/
├── frontend/
│   ├── src/
│   │   ├── components/     # Layout, Header, Sidebar, ProtectedRoute
│   │   ├── context/        # AuthContext
│   │   ├── pages/          # Dashboard, Inventory, Orders, Analytics...
│   │   ├── services/       # productAPI.js (axios)
│   │   └── hooks/
│   └── ...
│
├── backend/
│   └── app/
│       ├── routes/         # auth, product, orders, suppliers, analytics...
│       ├── ml/             # forecast_model.py (ARIMA)
│       ├── models.py       # SQLAlchemy models
│       ├── schemas.py      # Pydantic schemas
│       ├── database.py     # DB config
│       └── main.py         # FastAPI app
└── README.md

---

🚀 Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• Git

---

Backend Setup

# 1. Clone the repo
git clone https://github.com/your-username/ai-inventory-system.git
cd ai-inventory-system/backend

# 2. Install dependencies
pip install fastapi uvicorn sqlalchemy pydantic bcrypt PyJWT
pip install pandas numpy statsmodels

# 3. Create .env file

Create backend/.env:

JWT_SECRET=your-secret-key-here
SENDER_EMAIL=your-gmail@gmail.com
SENDER_PASSWORD=your-16-digit-app-password
APP_NAME=AI Inventory System
FRONTEND_URL=http://localhost:5173

# 4. Run migration (first time only)
python migrate.py

# 5. Start server
python -m uvicorn app.main:app --reload --port 8001

API docs: http://127.0.0.1:8001/docs

---

Frontend Setup

cd ../frontend

# 1. Install dependencies
npm install

# 2. Create .env file

Create frontend/.env:

VITE_API_URL=http://127.0.0.1:8001

# 3. Start dev server
npm run dev

App: http://localhost:5173

---

🔑 Default Login

Create your first admin user via API: POST /auth/create-user with { "email": "admin@test.com", "password": "123456", "role": "admin" }

---

📸 Screenshots

Dashboard	Analytics

Inventory	AI Assistant

---

🌍 Deployment

Frontend → Vercel

cd frontend
npm run build
# Push to GitHub → Vercel auto-deploys

Backend → Render

1. New Web Service → Connect GitHub repo
2. Root directory: backend
3. Build command: pip install -r requirements.txt
4. Start command: uvicorn app.main:app --host 0.0.0.0 --port 10000
5. Add environment variables from .env

---

📋 API Endpoints

Method	Endpoint	Description
POST	/auth/login	Login with JWT
POST	/auth/forgot-password	Send OTP
POST	/auth/verify-otp	Verify OTP
POST	/auth/reset-password	Reset password
GET	/products/	Get all products
POST	/products/	Create product
GET	/analytics/	Full analytics
GET	/forecast/{id}	Product forecast
GET	/notifications/	Get alerts
GET	/ai/ask?question=	AI assistant

---

👨‍💻 Author

Amit Sharma
📧 sharmaamit55174@gmail.com
🔗 LinkedIn
🐙 GitHub

---

📄 License

MIT License — feel free to use and modify.

---

Made with ❤️ using React + FastAPI