A Python Package for Distance-Based Generalized Sensitivity Analysis (DGSA)

Distance-based Generalized Sensitivity Analysis (DGSA) is a global sensitivity analysis method that quantifies single parameter sensitivities as well as two-way parameter interactions. One of the main advantages is that it can provide sensitivity analyses where the amount of model evaluation needed is limited, compared to other methods. A Python package is implemented to compute both single parameter and conditional parameter sensitivities using two types of sensitivity measures. Sensitivity results can be visualized using Pareto plots and heatmaps with a perceptually uniform colormap. The source code is stored in a GitHub repository and includes an example dataset to illustrate its usage. This tool helps provide more complete and deeper insights into parameter sensitivity analysis.