Propagating input uncertainties into parameter uncertainties and model prediction uncertainties—A review

A review of uncertainty quantification techniques is provided for a variety situations involving uncertainties in model inputs (independent variables). The interest are divided into three categories: (i) when prediction quantified based on uncertain inputs, (ii) parameter estimate calculated by propagation from measured and outputs, (iii) corresponding estimates. For all situations, linearization-based Monte Carlo-based reviewed details their algorithms presented. Recommendations which best different types chemical engineering models the amount input nonlinearity over range plausible values.