Multi-status Bayesian network for analyzing collapse risk of excavation construction

In response to the need for effective collapse risk analysis in engineering projects, a decision support approach is presented. It is rooted in multi-status Bayesian network (MSBN) and fuzzy set theory, encompassing MSBN construction, risk analysis, and management. This research addresses the causative correlation between influential factors and excavation collapse, considering the inherent uncertainty and fuzziness of the project. To enhance the reliability of evaluation results, an expert confidence index, integrating judgment ability, subjective reliability, and risk preference, is introduced. The approach is applied to an excavation case study, demonstrating its viability and capacity. Furthermore, it offers actionable insights for decision-makers to proactively mitigate accident probabilities. Notably, sensitivity analysis identifies critical risk factors. This research contributes to the field of engineering project risk assessment and management. It serves as a foundation for future research and development, guiding the way for improved strategies and decision support systems.