A shortcut model is a low-order model that provides a simplified representation of a complex process, thereby enabling faster and more efficient analysis, design and optimization without the computational intensity of a full-scale model. For continuous distillation columns, the Kremser-Edmister approach allows building reliable shortcut models by defining the effective absorption and stripping factors as average values between the top and bottom trays of the column. This paper shows that, unfortunately, this approach does not extend to batch distillation because the effective factors computed from the top and bottom trays can be quite inaccurate. Next, the paper proposes a novel column segmentation and a new approach for computing pertinent effective absorption and stripping factors. The resulting shortcut model is shown to accurately reproduce the dynamic behavior of a rigorous tray-by-tray model and to significantly reduce the computational time necessary to solve a minimal-time optimization problem subject to recovery and purity constraints.