Startup of distillation columns is one of the most difficult operations in the chemical industry. Since the startup often lasts a long period of time, leads to off-spec products, and costs much energy, optimisation of startup operating policies for distillation columns is of great interest. In the last few years, we have carried out both theoretical studies and experimental verifications with the purpose of minimising the startup time of distillation columns. Model-based optimisation as well as real plant implementation is the core of this work. The model is at first validated with experimental data and then used in the optimisation problem formulation. A rigorous column model and a sequential dynamic optimisation approach have been applied to several pilot distillation columns. Experimental results indicate that a significant reduction of the startup time can be achieved by implementing the developed optimal policies. These results demonstrate the applicability of the modelling and optimisation methodology. (C) 2003 Elsevier Ltd. All rights reserved.