The optimal operation of batch distillation systems has been studied in this paper. A new approach termed the limiting gradient approach is introduced for effectively identifying feasible and cost-effective operating profiles for batch distillation systems. A new semirigorous model for batch distillation has been presented for considering holdup of the column, which is applicable for different batch distillation configurations. A novel optimization model incorporating the limiting gradient approach has been proposed, which significantly reduces computational burden to search for a feasible operation of a batch distillation column by considerably reducing effort for carrying out a large number of dynamic simulations required for optimization. Case studies have been presented to demonstrate the applicability of the approach for various batch distillation configurations and to illustrate how the proposed optimization framework systematically addresses different performance indices and objectives.