Markov processes for maintenance optimization of civil infrastructure in the Netherlands

Abstract

The Netherlands, like many countries in this world, face a challenging task in managing civil infrastructures. The management of vital infrastructures, like road bridges, is necessary to ensure their safe and reliable functioning. The Directorate-General for Public Works and Water Management in the Netherlands manages the structures in the national road network. A large number of bridges and viaducts were constructed during the 1960's and 1970's. Due to many factors, it is difficult to determine the exact length of the remaining life of a structure. This is why the manager frequently performs inspections and registers the state of each structure in a database. A principal element of bridge management systems is the estimation of the uncertain rate of deterioration. This is usually done by using a suitable model and by using information gathered on-site during inspections. This thesis proposes a statistical and probabilistic framework, which enables the decision maker to estimate the rate of deterioration and to quantify his uncertainty about this estimate. The framework consists of a continuous-time Markov process with a finite number of states to model the uncertain rate at which the quality of structures reduces over time. The result of this research is a unified approach to modeling uncertain deterioration and decision making for optimal maintenance management. It has been succesfully applied to condition data of more than 3000 concrete structures in the Netherlands which were gathered from 1985 to 2004.