Data and code for "An explainable simulation–machine learning framework for target-driven early design of indoor ice arenas"

The package includes: the design dataset with its training/test split; the envelope options and unit costs; the trained XGBoost surrogates; SHAP values for all designs, global importance, interactions, non-linear breakpoints and cost slopes; the decision ranges (maximum admissible clear height and envelope bounds); 28 additional validation simulations.