Probabilistic service life estimation of roofing systems

The service life of building components is now considered as one of the major concerns in the construction industry given its effect on function, maintenance policy, life cycle cost and environmental impact. The main objective of this study is to propose a probabilistic factorial method enabling facility managers to estimate the service life of roofing systems. The estimation model proposed is primarily based on the factor method defined in ISO 15686-1, but is modified to include only factors relevant to for roofing systems: factor A (component quality), factor B (design level), factor C (work execution level), factor E (outdoor environment), factor F (in-use condition) and factor G (maintenance level). Forty-five roof cases are collected for analysis. For each case, the values of these factors are based on the Delphi method, in which a panel of experts is requested to assess each factor on the basis of a 7-point scale. Chi-square statistic shows that the best theoretical distributions for factors A, B, C, E, F and G are weibull, normal, weibull, logistic, weibull and lognormal distributions, respectively. The mean value and standard deviation of the service life of the roofing systems generated by the Monte Carlo simulation are 16.4 and 2.7 years, respectively. There is only a 5% chance that the estimated service life will be less than 12.5 years. This provides a reasonable estimate of a maintenance-free period and enables facility managers to formulate an optimal maintenance plan.