Optimum life-cycle maintenance strategies of deteriorating highway bridges subject to seismic hazard by a hybrid Markov decision process model

Highway bridges are a class of important infrastructure components in transportation networks. Highway bridges will inevitably deteriorate over time due to various external impacts, such as traffic loads and corrosion. In seismic-prone areas, seismic hazard is another critical factor that threatens the normal operation of ageing highway bridges. To ensure adequate safety of ageing highway bridges under potential earthquake shocks, it is imperative to perform proper maintenances during their service life. This paper aims at optimizing the life-cycle maintenance strategies for highway bridges under the combinatorial threats of progressive deterioration and sudden earthquakes. Driven by the distinct characteristics of the two threats, the authors have proposed a novel hybrid Markov decision process model that integrates a discrete-time Markov decision process model for dealing with progressive deterioration and a continuous-time Markov decision process model for dealing with sudden earthquakes into a unified framework. The proposed model is further reduced into the form of a discrete-time Markov decision process model with an equivalent state transition matrix and an equivalent discount factor, which can be easily solved by dynamic programming. A ternary state space that collectively describes corrosion initiation, corrosion damage cumulation, and seismic damage cumulation is defined. A gamma process and a Poisson process are adopted to model bridge corrosion and earthquake occurrences, respectively. To demonstrate the application of the proposed framework, a numerical example is conducted, in which the optimizations with respect to the maintenance strategy, inspection interval, and design performance level of a hypothetical highway bridge are carried out successively.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01759107
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Nov 19 2020 3:13PM