Zhang, LM (Zhang, Limao);Wu, XG (Wu, Xianguo); Ding, LY (Ding, Lieyun); Skibniewski, MJ (Skibniewski,Miroslaw J.); Yan, Y (Yan, Y.)

下载地址：

http://www.sciencedirect.com/science/article/pii/S0957417412012328

Abstract：This paperpresents a novel and systemic decision support model based on Bayesian Networks(BN) for safety control in dynamic complex project environments, which shouldgo through the following three sections. At first, priori expert knowledge isintegrated with training data in model design, aiming to improve theadaptability and practicability of model outcome. Then two indicators, ModelBias and Model Accuracy, are proposed to assess the effectiveness of BN inmodel validation, ensuring the model predictions are not significantlydifferent from the actual observations. Finally we extend the safety controlprocess to the entire life cycle of risk-prone events in model application,rather than restricted to pre-accident control, but during-constructioncontinuous and post-accident control are included. Adapting its reasoningfeatures, including forward reasoning, importance analysis and backgroundreasoning, decision makers are provided with systematic and effective supportfor safety control in the overall work process. A frequent safety problem,ground settlement during Wuhan Changjiang Metro Shield Tunnel Construction(WCMSTC), is taken as a case study. Results demonstrate the feasibility of BNmodel, as well as its application potential. The proposed model can be used bypractitioners in the industry as a decision support tool to increase thelikelihood of a successful project in complex environments.

Keywords: Decision support analysis; Safety control;Bayesian Networks (BNs); Ground settlement; Complex environments