An Integrated Quantitative Risk Assessment Method for Underground Engineering Fires

Area of Science:

• Engineering Safety
• Risk Assessment
• Disaster Management

Background:

• Underground engineering fires (UEFs) pose significant risks.
• Existing risk assessment methods may not fully capture the complexities of UEFs.

Purpose of the Study:

• To propose a novel risk assessment method for underground engineering fires (UEFs).
• To enhance the evaluation of UEF risks and support emergency management.

Main Methods:

• Developed an Event-Based Probability (EPB) model for UEFs, transforming it into a Fuzzy Bayesian Network (FBN).
• Utilized fuzzy numbers and interval probabilities for node states and conditional probabilities.
• Employed fuzzy Bayesian inference and α-weighted de-fuzzification to identify key risk factors and calculate sensitivity.

Main Results:

• The EPB-FBN model successfully identified key risk factors and maximum risk chains in analyzed scenarios.
• The method provided realistic analysis for fire safety evaluation.
• Demonstrated effectiveness in deductive analysis of three UEF scenarios.

Conclusions:

• The proposed EPB-FBN method offers a comprehensive approach to UEF risk assessment.
• This contributes a new perspective to UEF accident analysis and safety research.
• The model supports improved fire safety evaluation and emergency management in underground engineering.