A Dynamic Assessment Model for Managing Risk in Building Construction: Utilizing T-S Fuzzy Fault Tree and Bayesian Network Approach

The study introduces a dynamic evaluation framework that harmoniously combines T-S fuzzy fault tree and Bayesian network methodologies, with the aim of proficiently appraising and mitigating hazards related to construction safety. The T-S fuzzy fault tree model allows for the quantification of event risk probability by incorporating fuzzy theory. This enables a more accurate calculation of the probability of system accidents based on the risk status of fundamental events. By applying the conditional probability of the Bayesian network, a Bayesian network transformation model of the T-S fuzzy fault tree is constructed. The bi-directional inference algorithm of the Bayesian network facilitates the calculation of the probability of failure. The results of the study demonstrate that the proposed model provides quantitative calculations based on real construction scenarios. It helps in dynamically analyzing the probability of system accidents and troubleshooting fault occurrences, contributing to a more scientific and sensible assessment of building construction risks. Overall, this model provides a theoretical foundation for the assessment and precise management of safety risks in building construction. It can assist in improving construction safety by identifying potential risks, implementing preventive measures, and enhancing safety control measures.