Bayesian network-based models for bridge network management

Abstract : Maintenance for highway bridges is crucial in order to keep the network in a satisfactory condi-tion for users but is also a costly affair. This paper proposes a dynamic, Bayesian network-based model to provide cost-efficient strategies in the context of bridge network management. Characteristics related to un-certainties in both the degradation phase and subsequent maintenance strategies are handled through the de-sirable probabilistic dependencies properties BNs possess. The extension to a specific version of Influence di-agrams allows formulating the optimization part of the problem in order to eventually provide long-term strategies as well as minimize expected costs. To that end, a case study that tackles both conditional and un-conditional cases is presented.
Complete list of metadatas
Contributor : Alex Kosgodagan-Dalla Torre <>
Submitted on : Tuesday, May 2, 2017 - 5:45:48 PM
Last modification on : Friday, February 21, 2020 - 1:42:59 PM
Long-term archiving on: Thursday, August 3, 2017 - 1:58:34 PM


Files produced by the author(s)


  • HAL Id : hal-01517168, version 1


Alex Kosgodagan-Dalla Torre, Oswaldo Morales-Nápoles, Johan Maljaars, Bruno Castanier, Thomas Yeung. Bayesian network-based models for bridge network management. 25th European Safety and Reliability Conference, Sep 2015, Zürich, Switzerland. ⟨hal-01517168⟩



Record views


Files downloads