Abstract: Highlights•A learning process to estimate UFLS, focusing on common conventional schemes.•Feature selection and a new partitioning method for regression trees are analyzed.•Novel UFLS estimation with MILP models provides unique insights.
Loading