Probabilistic methods in management and inspection of buried steel pipe bridges

Abstract

This thesis studies how probabilistic methods and analyses can support management and inspection of bridges. The methods are displayed through a case study of buried steel pipe bridges. Bayesian networks, influence diagrams, reliability analyses, Markovian processes, regression analyses and cost, risk and benefit analyses are studied. The topics of consideration are decision making for inspectors, condition of buried steel pipe bridges, deterioration predictions, failure predictions and replacement strategies.

Management and inspection of bridges can benefit from the use of probabilistic analyses. The expert’s experience and knowledge, physical theory and probability theory are all important elements in optimal management. Bayesian networks and influence diagrams can be applied to several topics related to inspection and management of structures, and these methods may contribute to optimal decisions.