When you have completed a set of experiments and want to visualize how two different factors influence your outcome, use a response surface contour plot. This tool creates a 2D map that helps you identify the specific combination of settings that achieve your target result while holding other variables constant.
Visualizing Process Interactions
A response surface contour plot translates complex mathematical models into intuitive 2D maps. By projecting a multi-dimensional experiment onto a flat surface, you can immediately see the relationship between two inputs and a single output. This makes it easier to communicate findings and identify where your process is most stable.
The plot uses your previously fitted model to calculate predictions across a grid. The resulting contours act like a topographic map, where closely spaced lines indicate regions where your response changes rapidly as you adjust your process settings.
Navigating the Design Space
When your experiment involves more than two factors, this tool acts as a lens. You can select any pair of variables to visualize while keeping the others fixed at the experiment's center. This functionality allows you to explore the entire experimental space systematically, ensuring you do not miss optimal configurations hidden in higher dimensions.
The inclusion of training points on the plot provides a visual reference to your original data. This helps you verify that your model's predictions are well-supported by actual measurements, rather than being an artifact of mathematical extrapolation.
Data-Driven Decision Making
Relying on a model's goodness-of-fit metrics is important, but spatial visualization is how you turn numbers into actionable process parameters. By identifying the 'sweet spot' on the contour map, you can confidently set your machine or process parameters to achieve the desired output.
Because this tool is integrated into the Lattice environment, it works directly with your existing experimental data. As you update your model or add new runs, the contour visuals adjust automatically, providing a consistent way to monitor your progress toward an optimized process.
1 · Intent → method
An LLM picks rsm_plot_contour from a fixed catalog.
2 · Method → numbers
Deterministic Python engine runs the math. Same input → same output.
3 · Numbers → plain language
A second LLM translates the result into your domain’s vocabulary.
Why is the plot showing different values when I change the fixed factors?
A response surface contour plot visualizes two factors at a time. Because your model might include three or more factors, Lattice fixes the remaining factors at their center points. Changing these fixed values shifts the slice of the model you are looking at.
What do the colored regions in the plot represent?
The colors represent different levels of your response variable, predicted by your fitted model. By following the gradients, you can identify which areas of your factor settings lead to higher or lower outcomes, helping you narrow down your target operating window.
Tool input schema
Schema for rsm_plot_contour not exported yet (run pnpm export:registry).