Predictive worker safety assessment through on-site correspondence using multi-layer fuzzy logic in outdoor...

Area of Science:

• Construction Safety Engineering
• Artificial Intelligence in Civil Engineering
• Internet of Things (IoT) for Workplace Safety

Background:

• Construction sites present significant global safety risks, leading to injuries and fatalities.
• Existing quantitative safety solutions lack accuracy, hindering widespread adoption.
• There is a critical need for efficient fuzzy inference systems to improve construction safety.

Purpose of the Study:

• To propose an edge inference framework utilizing multi-layered fuzzy logic for construction worker safety.
• To develop a system for real-time safety risk assessment and monitoring.
• To enhance proactive safety measures and reduce workplace hazards.

Main Methods:

• An edge computing framework with IoT devices for data collection, storage, and management.
• Multi-layer fuzzy logic applied to infer a worker safety index based on environmental and worker data.
• Integration of weather, building condition (load, strain, inclination), and worker biometrics (heart rate, location) as inputs.

Main Results:

• The framework infers an integrated safety index using a multi-layered fuzzy logic approach.
• The system provides real-time warnings and error measurements on a safety scale (1-10).
• A web service enables real-time monitoring of worker safety and IoT device status.

Conclusions:

• The developed framework demonstrates efficiency in risk assessment, real-time monitoring, and proactive safety actions.
• The system contributes to a safer and more productive construction work environment.
• Early detection of sensor malfunctions ensures continuous worker safety assurance.