Stochastic Models of Uncertainties in Computational Mechanics presents the main concepts, formulations, and recent advances in the use of a mathematical-mechanical modelling process to predict the responses of a real structural system in its environment. Computational models are subject to two types of uncertainties—variabilities in the real system and uncertainties in the model itself—so, to be effective, these models must support robust optimization, design, and updating.

A probabilistic approach to uncertainties is the most powerful, efficient, and effective tool for computational modeling. Chapters describe the methodology, construction, and estimation for using parametric, nonparametric, and generalised probabilistic approaches to the uncertainties in computational structural dynamics. Other chapters demonstrate the nonparametric probabilistic approach in the context of linear and nonlinear structural dynamics and in structural acoustics and vibration. A new methodology adapted to partial and limited experimental data is presented to identify random coefficients in high-dimension polynomial chaos expansions.

This compact guide to probabilistic approaches in computational modelling will be of interest to those working in computational sciences, especially structural engineers, who are developing and applying stochastic models of uncertainties to predict real-world structural systems.