Unlike one-fraction batch distillation using constant distillate composition control, a priori policies are to be tracked with control systems designed to realize the optimal operation of multiple-fraction batch distillation. Adaptive control is proposed to carry out this task. Based on a detailed model, characteristics of batch distillation control are analysed and a proper system is designed for an existing multicomponent/multiple-fraction batch column. Besides tracking the optimal reflux ratio profile, the maximum vapour load must be maintained during the batch to realize the optimal operation of multiple-fraction batch distillation. In addition, a changing temperature profile of the condenser should be followed to reduce the operating energy with a possibly minimum subcooling. Recursive least square estimation (RLSE) with a variable forgetting factor is applied to the on-line identification of the plant to follow the changing dynamics of the process. Generalized predictive control (GPC) is successfully used to track the predefined policies. Moreover, we show that it is difficult to control the process satisfactorily over the whole batch time by PID controllers with fixed parameters.