Non Deterministic Approaches

Robustness is the degree to which a system or a component can function correctly in the presence of invalid inputs or stressful environment condition.

Robustness analysis aims to estimate the sensitivity of outputs to the inputs variability. Standard deviation $\sigma$ is used as measurement of the robustness of the outputs. The smaller the standard deviation of an output is, the more robustness the output is. Based on the prediction of the sensitivity of outputs, optimization methods can be applied to improve the robustness of the outputs.

There are two methods to calculate the standard deviation of an output: sampling method and approximation method. The sampling method is essentially a Monte Carlo simulation, while the approximation method is represented by the First Order Second Moment (FOSM) and by a Monte Carlo simulation on a linear model.