PROBABILISTIC MODELS FOR THE ANALYSIS AND PREDITION OF THE DURABILITY OF ARCHITECTURAL STRUCTURES

Abstract

This paper examines the application of probabilistic models for the analysis and prediction of the durability of architectural structures under uncertainty in material properties and operational conditions. The research methodology is based on reliability theory and time-to-failure analysis, with particular emphasis on the Weibull and lognormal distributions, which are widely used to describe degradation processes in building materials and structural systems. The study employs synthetic datasets generated on the basis of empirical results reported in previous investigations of the durability of concrete and structural materials. Structural reliability methods and Monte Carlo simulation are applied to estimate failure probabilities and to predict the service life of architectural objects. The results demonstrate the effectiveness of the probabilistic approach for quantitative assessment of degradation risks and enable the estimation of expected service life while accounting for the variability of influencing factors. The proposed methodology can be used as a decision-support tool in the design process and in the reliability assessment of architectural structures.