🏭 Smart Manufacturing IoT‑Cloud Monitoring Dashboard

A smart manufacturing monitoring dashboard powered by IoT sensor data.

Real-time machine health tracking, anomaly detection, and maintenance risk analysis — all in one place.

---

📌 Table of Contents

• Overview
• Dashboard Preview
• Tech Stack
• Dataset
• Project Structure
• Features
• Getting Started
• Future Improvements
• Author

---

🧠 Overview

In modern smart factories, machines generate continuous streams of sensor data. Without proper monitoring, anomalies go unnoticed and unexpected breakdowns cost time and money. This project tackles that problem by building an interactive monitoring dashboard using historical IoT sensor data from multiple manufacturing machines.

Key highlights:

• Live machine-wise sensor trend visualization
• Anomaly summary across all machines
• Recent health flag table per machine
• Rule-based maintenance risk checker (demo)
• Optional ML model integration (RandomForest) for predictive maintenance

---

📸 Dashboard Preview

Add a screenshot of your running Streamlit dashboard here.

![Dashboard Screenshot](assets/dashboard_screenshot.png)

---

🛠 Tech Stack

Category	Tools
Language	Python 3.x
Dashboard UI	Streamlit
Data Handling	pandas, numpy
Visualization	matplotlib
ML (Optional)	scikit-learn (RandomForest)
Model Persistence	pickle
Environment	Jupyter Notebook, VS Code

---

📂 Dataset

File: smart_manufacturing_data.csv

Column	Description
timestamp	Date and time of the reading
machine_id	Unique machine identifier
temperature	Machine temperature
vibration	Vibration level
humidity	Ambient humidity
pressure	System pressure
energy_consumption	Energy usage during interval
anomaly_flag	0 = Normal, 1 = Anomaly detected
failure_type	Type of failure (Normal, Overheating, Vibration Issue, Pressure Drop, etc.)
predicted_remaining_life	Estimated remaining useful life
downtime_risk	Continuous risk score for possible downtime
maintenance_required	0 = No maintenance, 1 = Maintenance needed

---

🗂 Project Structure

smart-manufacturing-iot-monitoring/
│
├── 🚀 app.py                        # Streamlit dashboard (main app)
├── 📊 smart_manufacturing_data.csv  # IoT manufacturing dataset
├── 📓 PROJECT.ipynb                 # Jupyter notebook for EDA & experiments
├── 🤖 rf2_model.pkl                 # (Optional) Saved RandomForest model
├── ⚖️  scaler.pkl                   # (Optional) Saved feature scaler
└── 📄 README.md                     # Project documentation

---

✨ Features

1. 📈 Machine-wise Sensor Monitoring

• Sidebar lets you select any machine_id
• Time-series line chart of temperature, vibration, and pressure
• Quickly spot abnormal spikes, trends, and outliers

2. 🏥 Recent Health Flags

• Table showing the latest records for the selected machine
• Columns: timestamp, anomaly_flag, maintenance_required, downtime_risk
• Instant snapshot of machine health status

3. ⚠️ Overall Anomaly Summary

• Aggregates anomaly counts across all machines
• Ranked table of top machines by anomaly frequency
• Helps maintenance teams prioritize inspections

4. 🔧 Simple Maintenance Risk Check (Demo)

• Manually adjust Temperature, Vibration, and Pressure using sliders
• Rule-based logic compares values to historical quantiles
• Returns risk classification:

Risk Level	Meaning
🟢 LOW	Maintenance not urgent
🟡 MEDIUM	Keep machine under observation
🔴 HIGH	Maintenance recommended immediately

This demo logic can be replaced by the trained ML model (rf2_model.pkl) for production use.

---

🚀 Getting Started

1. Clone the Repository

git clone https://github.com/Musawir456/smart-manufacturing-iot-monitoring.git
cd smart-manufacturing-iot-monitoring

2. (Optional) Create a Virtual Environment

python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Run the Streamlit Dashboard

streamlit run app.py

5. Open in Browser

http://localhost:8501

---

🔮 Future Improvements

• 🤖 ML Integration — Replace rule-based logic with the trained rf2_model.pkl RandomForest model for real predictions
• 📡 Real-time Streaming — Connect live IoT streams via MQTT or HTTP instead of static CSV
• 🔐 Authentication — Add role-based access for operators, engineers, and managers
• ☁️ Cloud Deployment — Deploy Streamlit app to AWS / GCP / Streamlit Cloud for remote access
• 📧 Alert System — Automated email/SMS alerts when HIGH risk is detected

---

🗃 Optional ML Artifacts

File	Description
rf2_model.pkl	Trained RandomForest model to predict maintenance_required from sensor data
scaler.pkl	Feature scaler used during model training

These artifacts are trained and ready — they can be loaded into app.py for real-time predictive maintenance with minimal code changes.

---

🖼️ Screenshots

Dashboard

Upload Page

Page

Results

---

👨‍💻 Author

Abdul Musawir BS IT / Computer Scienc AI/ML Engineer & Data Scientist 🎓 The Superior University, Lahore 📧 abdulmusawir8191456@gmail.com

---

⭐ If you found this project useful, please give it a star! ⭐

Made with ❤️ by Abdul Musawir — The Superior University Lahore