Assessment Measures in Uncertainty Propagation Analysis Applied to Composite Structures

The uncertainty propagation of composite structures is investigated considering descriptive statistical measures of the response variability. A study based on sensitivity to uncertainty that allows selecting the important parameters using sensitivity indices is presented. The uncertainty propagation and the importance measure of input parameters are analysed using four different approaches: a first order local method, a global sensitivity analysis supported by a variance-based method, an extension of local variance to estimate the global variance over the domain of inputs and an Artificial Neural Network-based Monte Carlo approach. Advantages of different approaches are discussed by comparing the obtained results.