fig11

From: Linking processing parameters with melt pool properties of multiple nickel-based superalloys via high-dimensional Gaussian process regression

Linking processing parameters with melt pool properties of multiple nickel-based superalloys via high-dimensional Gaussian process regression

Figure 11. Sensitivity Analysis for Linking processing parameters with melt pool properties of multiple nickel-based superalloys via high-dimensional Gaussian process regression . (A) The first-order and total-order Sobol indices presented as bar graphs for each uncertainty parameter (input variable); (B) second-order Sobol indices as a heat map; (C) absolute mean SHAP plot where aggregate SHAP values for each uncertainty parameter are plotted in descending order; (D) summary of SHAP analysis presented as a beeswarm plot. Each point on the summary plot corresponds to a Shapley value for an input variable and an instance, where the input variables are provided on the y-axis and the Shapely values along the x-axis. For each group, the color of the points is determined by the value of the same uncertainty parameters. The uncertainty parameters are ordered by the mean SHAP values.