Study on a multi-factor lane-changing risk resilience assessment model based on genetic algorithm and fault tree...

Area of Science:

• Intelligent Transportation Systems
• Road Safety Engineering
• Resilience Engineering

Background:

• Existing lane-change risk models struggle with dynamic weather adaptation and real-world validation.
• A resilience engineering approach is needed to assess a system's capacity to maintain safety under disturbances.

Purpose of the Study:

• To develop a weather-aware, resilience-oriented lane-change risk assessment framework.
• To introduce novel metrics, Risk Exposure Level (REL) and Risk Severity Level (RSL).
• To improve dynamic risk adaptation for adverse weather conditions.

Main Methods:

• Integrated a Genetic Algorithm (GA)-calibrated Stopping Sight Distance (SSD) model with Fault Tree Analysis (FTA).
• Utilized the CitySim naturalistic driving dataset to extract lane-change events under various weather conditions.
• Employed statistical testing to identify key influencing factors on lane-change risk.

Main Results:

• Optimized braking parameters using GA for self-adaptive risk thresholds in different weather.
• Quantified conflict probability (REL) and severity (RSL) for comprehensive system robustness assessment.
• Achieved a 42.38% improvement in model fitness compared to the baseline.

Conclusions:

• The developed framework effectively captures real-time lane-changing risk, even in adverse weather.
• The model provides a reliable tool for proactive safety management in intelligent transportation systems.
• The resilience-oriented approach enhances decision support for intelligent transportation systems.