The ageing of power systems and their components are of great concern for the asset manager, especially in view of the coming replacement wave in the electrical infrastructure. The asset manager is in need for a prediction tool, to determine the long term effects on the statistical relevant parameters of a population of power systems components. In this paper we describe a method for forecasting the population reliability from the individual reliabilities. The individual reliabilities are based on the degradation of the transformer insulation paper. The prediction of the replacement wave is highly sensitive to its input parameters, such as historical load and future load growth.