Application of global sensitivity analysis for identification of probabilistic design spaces

The design space (DS) is defined as the combination of materials and process conditions that provides assurance of quality for a pharmaceutical product. A model-based approach to identify a probability-based DS requires costly simulations across the entire process parameter space (certain) and the uncertain model parameter space and material properties space. We demonstrate that application of global sensitivity analysis (GSA) can significantly reduce model complexity and reduce computational time for identifying and quantifying a probabilistic DS by screening out nonimportant uncertain parameters. The novelty of this approach is that the use of an indicator function, which takes only binary values as a model function, allows application of a straightforward GSA based on Sobol' sensitivity indices and avoids using more costly Monte Carlo filtering or GSA for constrained problems. We consider a chemical reaction example to illustrate how this formulation results in a model reduction and a significant decrease of model runs.