When CVaR Meets With Bluetooth PAN: A Physical Distancing System for COVID-19 Proactive Safety.

Area of Science:

• Computer Science
• Public Health
• Risk Management

Background:

• The COVID-19 pandemic highlighted the need for effective physical distancing strategies.
• Existing methods often lack robust risk assessment for ensuring personal safety.
• Bluetooth-enabled personal area networks (PANs) offer potential for proximity monitoring.

Purpose of the Study:

• To develop a risk-aware physical distancing system for pandemic scenarios.
• To formulate the physical distancing problem using Conditional Value-at-Risk (CVaR).
• To enhance personal safety by ensuring adequate private distances.

Main Methods:

• A two-stage approach combining control flow, linear modeling, and curve fitting.
• Probabilistic linear model to determine safe movement distance based on CVaR constraints.
• Levenberg-Marquardt (LM)-based curve fitting for optimal trajectory planning.

Main Results:

• The system effectively reduces the risk associated with recommending safety distances.
• Experimental validation using Bluetooth data confirms system effectiveness.
• Achieved a 45.11% risk reduction for safe movement distance calculation at 95% CVaR confidence.

Conclusions:

• The proposed risk-aware physical distancing system enhances private safety during pandemics.
• The CVaR-based formulation provides a robust method for tail-risk management.
• The integrated approach offers a practical solution for real-world deployment.