Smart Factory IoT Digital Twin
A real-time Streamlit dashboard simulating a smart factory with 50 IoT sensors, an autonomous robot, AI-driven anomaly detection, and interactive controls.
Features
Factory Floor Map
·	3 operational zones: Assembly (A), Logistics (B), Safety (C)
·	50 sensors (10 types × 5 each) plotted as color-coded nodes:
o	Zone A: Temp, Vibration, Current, Light
o	Zone B: Humidity, Ultrasonic, IR
o	Zone C: Pressure, Smoke, Fire
·	Autonomous robot (yellow diamond) navigates toward fault/override sensors at double speed, otherwise patrols random waypoints
·	Hover tooltips on sensor nodes (via Altair interactive chart below the map) showing ID, type, value, and status
Simulation Engine
·	Each sensor fluctuates with ±2% Gaussian noise around its baseline
·	Fault injection: Random (3 sensors) or ALL — sets override values at 1.6× or 2.0× baseline for 15 seconds
·	Sensors auto-detect ANOMALY status when value exceeds 1.3× baseline
·	Status color code: 🟢 NORMAL, 🟡 OVERRIDE, 🔴 ANOMALY
Controls (top bar)
·	Start/Stop simulation
·	AI toggle to enable/disable anomaly detection
·	Sidebar: Fault injection buttons (Random 3 / ALL)
Right Panel
1.	AI Analytics — real-time summaries: avg/max temperature, fault count, total alerts
2.	AI Alerts — last 6 alerts with severity color (HIGH = red, MEDIUM = yellow)
3.	Robot Teleport — X/Y sliders to reposition the robot instantly
4.	Sensor Override — select any sensor, set a custom value and duration
Sensor Data Table (below map)
·	Live snapshot of all 50 sensors: Timestamp, ID, Type, Zone, Value, Base, Dev %, Status
·	Interactive hover chart (Altair) — hover to see ID, type, value, status
·	Sensor details popup — select a sensor → popover with full details + history line chart
·	Download CSV (appears when stopped) — exports complete session history (all readings across all sensors)
AI Anomaly Detection
·	Threshold alert: value exceeds 130% of baseline → HIGH alert + toast notification + beep sound
·	Trend alert: monotonically rising values over 10 consecutive samples → MEDIUM alert
·	Sound (Windows): winsound.Beep(880Hz, 300ms) on HIGH alerts
Tech Stack
Component	Technology
Dashboard	Streamlit
Map rendering	Matplotlib
Interactive chart	Altair (shipped with Streamlit)
Data processing	Pandas, NumPy
Sound alerts	winsound (Windows)
MQTT bridge	paho-mqtt (optional, configured in config.py)

Sensors
Type	Base Value	Zone	Count
Temp	45	A	5
Vibration	2.5	A	5
Current	12	A	5
Light	500	A	5
Humidity	45	B	5
Ultrasonic	2.0	B	5
IR	1.0	B	5
Pressure	101.3	C	5
Smoke	10	C	5
Fire	5	C	5

Getting Started
cd smart_factory
pip install -r requirements.txt
streamlit run app.py

Open http://localhost:8501 in your browser.
Optional Dependencies
·	matplotlib — factory floor map with zone rectangles
·	winsound — alert beeps (Windows only, included in standard library)
Project Structure
.
├── smart_factory/
│   ├── app.py                 # Main Streamlit dashboard
│   ├── config.py              # MQTT broker settings
│   ├── sensor_sim.py          # Sensor simulation logic
│   ├── mqtt_publisher.py      # MQTT telemetry publisher
│   ├── ai_anomaly_detector.py # AI detection engine
│   ├── fault_injector.py      # Fault injection module
│   ├── dashboard.py           # Alternative dashboard
│   ├── main.py                # Entry point
│   ├── run_all.py             # Launch all components
│   └── requirements.txt       # Python dependencies
├── README.md
└── .gitignore

Usage Tips
1.	Click Start to begin simulation
2.	Inject faults via sidebar to see robot react
3.	Enable AI Detection for automated alerts
4.	Hover over the interactive scatter chart below the map to inspect sensors
5.	Use the details selector to view full sensor history
6.	Stop simulation and click Download CSV to export all session data