---

OptiStocks – Intelligent Inventory Optimization & Demand Forecasting 🚚📊

OptiStocks is a full-stack inventory management and demand forecasting platform designed for modern businesses to monitor inventory, analyze sales performance, and predict future demand using ARIMA/SARIMA models.

The system combines an intuitive web dashboard with a powerful time-series forecasting engine, helping companies reduce stockouts, avoid overstocking, and make data-driven purchasing decisions.

DEPLOYED SYSTEM: https://demand-forecasting-pink.vercel.app/ DEMONSTRATION VIDEO: https://drive.google.com/file/d/1Wk7GhwdC_-8pSQjVlj-MPzYl3weFVwKf/view?usp=sharing

---

✨ Key Features

• Add & manage companies and their stocks
• Real-time inventory analytics & KPIs
• ARIMA / SARIMA based demand forecasting
• Interactive visualizations for historical & forecasted sales
• Order recommendations dashboard
• Secure authentication (JWT-based)
• Deployed full-stack application with ML integration

---

🛠 Tech Stack

🔹 Frontend

🔹 Backend

🔹 Machine Learning / Analytics

---

🚀 Deployment

Component	Platform	Link
Frontend	Vercel	https://demand-forecasting-pink.vercel.app
Backend API	Render	https://optistocks-optimizer.onrender.com
ARIMA Forecasting (ML)	Streamlit Cloud	https://demandforecast1.streamlit.app

⚠️ Note: Render uses free instances.
The backend may take 5–10 seconds to wake up on first request.

---

🧪 Bonus & Evaluation Highlights

✅ Attempted Full-Stack Track (Web + API + Database)
✅ Implemented authentication using JWT
✅ Integrated Machine Learning pipeline (ARIMA/SARIMA)
✅ Separate ML deployment using Streamlit Cloud
✅ Deployed frontend & backend separately
✅ Focused on UX, dark theme, smooth navigation & visual clarity
✅ Clean project structure & documentation

---

🖥 Local Setup Instructions

1️⃣ Clone the repository

git clone https://github.com/sanjayjr8/Demand-Forecasting.git
cd Demand-Forecasting/optistocks_optimizer

2️⃣ Start Backend (Node + Express)

cd server
npm install

Create a .env file inside server/:

MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret
PORT=10000

Then run:

npm run dev

Backend runs at:

http://localhost:10000

3️⃣ Start Frontend (React + Vite)

Open a new terminal:

cd client
npm install
npm run dev

Frontend runs at:

http://localhost:5173

4️⃣ Run ARIMA Streamlit App (Optional – Local)

cd server/arima
pip install -r requirements.txt
streamlit run test.py

---

📸 Screenshots / Demo

---

📌 Assumptions

• Uploaded CSV data follows a time-series format (Month, Sales)
• Render free tier cold start delay is expected
• Forecasting parameters are user-controlled to demonstrate flexibility

---

👤 Author

Sanjay J
Final Year B.Tech – CSE
GitHub: https://github.com/sanjayjr8
LinkedIn: https://linkedin.com/in/sanjayj08

Built with focus on clean design, scalability, and real-world usability.