Condition Deterioration Prediction of Bridge Elements Using Dynamic Bayesian Networks (DBNs)
The ability of bridge deterioration models to predict future condition provides significant advantages in improving the effectiveness of maintenance decisions.This paper proposes a novel model using Dynamic Bayesian Networks (DBNs) for predicting the condition of bridge elements.The proposed model improves prediction results by being able to handle,deterioration dependencies among different bridge elements,the lack of full inspection histories,and joint considerations of both maintenance actions and environmental effects.With Bayesian updating capability,different types of data and information can be utilised as inputs.Expert knowledge can be used to deal with insufficient data as a starting point.The proposed model established a flexible basis for bridge systems deterioration modelling so that other models and Bayesian approaches can be further developed in one platform.A steel bridge main girder was chosen to validate the proposed model.
bridge deterioration models condition ratings dynamic Bayesian networks (DBNs) expert knowledge
Ruizi Wang Lin Ma Cheng Yan Joseph Mathew
Science and Engineering Faculty CRC for Infrastructure and Engineering Asset Management Queensland University of Technology 2 George Street, Brisbane QLD 4001, Australia
国际会议
成都
英文
564-569
2012-06-15（万方平台首次上网日期，不代表论文的发表时间）