Probabilistic updating of corroded reinforced concrete pipes for maintenance planning optimisation

Reinforced concrete pipes (RCP) are sometimes used to distribute aggressive water in industrial systems. Such networks of RCP deteriorate with time due to internal corrosion (Peter-Lazar 2000). Due to the low O-2 - content of aggressive water, a slow corrosion speed is expected for such applications. If the RCP in these networks are not periodically replaced, they will eventually fail. Strategies for replacement for these pipes depend on (1) the risks associated with failure of a water distribution network, and (2) the costs associated with replacing the pipes, including the cost of removing existing pipes, the cost of the new pipes and the cost of service interruption of there is any temporary closure (Luong 2002). Due to the lack of statistical data in regards with RCP failure, the development of a risk-based replacement strategy is not an easy task. This paper gives an example of how predictive models of the deterioration of RCP and the consequences of failure can be used to develop risk-based replacement strategies for RCP networks. Replacement strategies are determined for a series network consisting of 251 RCP. The main contribution of this paper is to provide a theoretical approach based on (1) the results of a reliability study, the use of (2) random drawing (binomial law) and (3) the decision tree method to determine optimal replacement strategy.