Industrial environments are highly prone to safety hazards due to the presence of harmful gases, abnormal temperature and humidity conditions, and mechanical vibrations. Continuous monitoring and early prediction of such parameters are essential to prevent accidents and ensure worker safety. This project presents a Smart Industrial Safety Monitoring System using Machine Learning, designed to monitor critical environmental conditions and provide real-time analysis and forecasting. The proposed system integrates multiple sensors, including a CO₂ sensor capable of measuring CO₂ concentration and Total Volatile Organic Compounds (TVOC), a DHT11 sensor for temperature and humidity, a gas sensor for detecting hazardous gases, and a vibration sensor for identifying abnormal mechanical activity. These sensors are interfaced with an Arduino Nano, which collects and preprocesses the sensor data. The processed data is then transmitted to a Raspberry Pi 3 A+, which acts as the central processing and communication unit. All sensor data is uploaded to the webpage for real-time storage, visualization, and remote monitoring. A local LCD display provides on-site visualization of the current environmental conditions. Machine learning algorithms are applied to the historical sensor data stored in the cloud to analyze patterns, detect anomalies, and forecast potential hazardous situations before they occur. By combining IoT, web monitoring and machine learning, the proposed system enhances industrial safety through continuous monitoring, intelligent data analysis, and predictive decision-making. This system is scalable, cost-effective, and suitable for deployment in various industrial environments to reduce risks, improve operational safety, and support proactive maintenance strategies.      

Objectives:

1. To monitor industrial environmental conditions such as CO₂, TVOC, temperature, humidity, gas leakage, and vibration.
2. To collect sensor data using Arduino Nano.
3. To send the collected data to Raspberry Pi for processing.
4. To upload all sensor data to the web for online monitoring.
5. To display real-time sensor values on an LCD screen.
6. To analyze sensor data using machine learning techniques.
7. To predict unsafe conditions in advance and improve industrial safety.

      The major building blocks of the project are:  

• Power supply.
• Raspberry pi3 A+.
• Arduino NANO.
• DHT11 (temperature and humidity) sensor.
• Vibration sensor.
• Gas Sensor.
• Co2 Sensor.
• LCD display.
• SD card.

      Software’s used:  

• Rasppbian OS.
• Machine learning algorithm.
• Express SCH for Circuit design.

      Block diagram:    

video: