Probabilistic methods for service life estimation of civil engineering structures

This chapter describes the background to the estimation of service life for use in lifecycle cost analyses of infrastructure, including deteriorating infrastructure. Since loadings, material properties and dimensions are not known with certainty, service life estimates must be based on probabilistic analyses. The theory most useful for this, the theory of structural reliability, is outlined briefly and related to service life estimation through stochastic process theory. This requires the estimation of the probability of failure on first load application and the so-called ‘out-crossing rate’. Methods to estimate these include Monte Carlo methods and moment methods. Both are revised briefly and future trends for computational methods indicated.