Coupling analysis of crane accident risks based on Bayesian network and the N-K model.

Area of Science:

• Construction Safety Engineering
• Risk Management
• Accident Analysis

Background:

• Crane operations are vital in construction but carry high accident risks.
• Effective accident prevention requires understanding complex interactions between risk factors.
• Existing methods may not fully capture the interplay of human, machine, management, and environmental influences.

Purpose of the Study:

• To propose a quantitative coupling method for analyzing crane accident risks.
• To investigate the interaction among human, machine, management, and environmental factors.
• To enhance the safety of lifting operations through improved risk assessment.

Main Methods:

• Utilizing text mining and fault tree analysis to identify and categorize risk factors.
• Developing a Bayesian network (BN) model integrated with the N-K model for risk coupling analysis.
• Defining types of risk coupling and quantifying their impact.
• Applying posterior probability and sensitivity analysis to assess failure probabilities.

Main Results:

• Identified that human and management factors significantly influence crane accidents.
• Demonstrated that the interplay among multiple risk factors is crucial in determining accident probability.
• Quantified the coupling effects of various risk factors using the BN and N-K models.

Conclusions:

• Human and management factors require heightened attention in crane safety protocols.
• The proposed quantitative coupling method provides a robust framework for analyzing complex crane accident risks.
• Comprehensive understanding of risk factor interactions is essential for effective accident prevention in construction.