Bayesian-network-based Fall Risk Assessment for Steel Construction Projects

DOI: 10.5176/2301-394X_ACE14.53

Authors: Sou-Sen Leu, Ching-Miao Chang, Jun-Yang Shi

Abstract:

Steel structure has been the most common technology used in the high-rise buildings. However, fall (or called as tumble) is the first occupational accident in the steel construction (SC) projects due to work at a height. In Taiwan, the percentage of fall in the SC projects is up to 67{6e6090cdd558c53a8bc18225ef4499fead9160abd3419ad4f137e902b483c465} over the past decade (2000-2010). Particularly for the steel member lifting work, due to unqualified safety equipment and unsafe worker behaviors, fall accidents at the SC sites constantly occur. To analyze the safety risk, several systematic safety risk assessment approaches, such as Fault Tree Analysis (FTA) and Failure Mode and Effect Criticality Analysis (FMECA), have been used to evaluate the safety risk at the SC projects. Nevertheless, the classical approaches ineffectively address dependencies among safety factors at different levels, which fail to provide early warning to prevent the occupational accidents. To overcome the limitations of the traditional approach, this paper discussed the development of a fall risk assessment model for the SC projects by establishing a Bayesian Network (BN) based on Fault Tree (FT) transformation. The model was proved to gain much better site safety management ability by better understanding of the probability of fall risks through the analysis of fall causes and their relationships in BN. The system has been used to analyze and verify against six SC projects currently under construction in Taiwan. It was found that the BN analysis is consistent with the conventional safety performance assessment. In practice, based upon the analysis of BN by inputting prior information about basic safety causes, the probabilities of fall risks are effectively assessed. Proper preventive safety management strategies can then be established to reduce the occurrences of fall accidents at the SC projects.

Keywords: Bayesian network, fault Tree, steel construction, fall risks