K-Means Clustering-Based Safety System in Large-Scale Industrial Site Using Industrial Wireless Sensor Networks.

Area of Science:

• Industrial safety engineering
• Wireless communication networks
• Cluster analysis

Background:

• Industrial sites utilize numerous workers and heavy equipment, making accident prevention critical.
• Existing safety systems are inadequate for large-scale sites where workers are dispersed over large areas.
• Current methods struggle with timely threat notification and exhibit high energy consumption in expansive industrial environments.

Purpose of the Study:

• To propose an energy-efficient safety system for large-scale industrial sites.
• To address the limitations of existing safety systems in handling dispersed workforces.
• To enable timely threat notification in expansive industrial settings.

Main Methods:

• Implementation of a k-means clustering algorithm to group workers across large areas.
• Utilizing a multicasting tree for threat notification delivery to each cluster.
• Employing local flooding within clusters for efficient notification dissemination.

Main Results:

• The proposed system effectively delivers threat notifications within a valid time frame.
• Demonstrated significant energy efficiency compared to existing safety schemes.
• Successfully groups dispersed workers for targeted communication.

Conclusions:

• The k-means clustering-based safety system is suitable for large-scale industrial environments.
• The system enhances safety by providing timely threat alerts.
• The proposed approach offers a more energy-efficient solution for industrial safety communication.