How to Perform a Process Parameter Screening

A screening study systematically varies process parameters across a defined range to map the design space. This is how you identify which parameters matter most and find operating conditions that are robust to normal process variability.

Overview

Screening studies answer questions like: what happens if the flow rate drifts by 10%? How sensitive is purity to the gradient slope? Is there a region of the design space where both yield and purity meet specification with comfortable margins?

In a traditional lab setting, answering these questions requires a large design of experiments (DOE) — often 20–50+ runs. With a calibrated mechanistic model, the same design space can be mapped with thousands of simulations in a fraction of the time and at no material cost.

A typical screening study involves:

1. Selecting the process or model parameters to screen and defining realistic ranges for each — based on expected process variability or the design space you want to explore.
2. Configuring the study — choosing the number of levels per parameter and the sampling strategy (grid, random, or Latin hypercube).
3. Running the simulations.
4. Analyzing the results — looking at how each parameter affects the key performance indicators (yield, purity, productivity) and identifying parameter interactions.
5. Defining robust operating conditions — regions where performance meets specification even when parameters vary within their expected ranges.

Screening studies are particularly valuable for process characterization and regulatory submissions, where you need to demonstrate understanding of the design space and justify proven acceptable ranges for critical process parameters.

Selecting Parameters and Ranges

Configuring the Screening Study

Running the Study

Interpreting the Results

Identifying Robust Operating Conditions