Separation processes account for more than half of capital and operating costs in chemical manufacturing. Separations are the energy intensive operations in the chemical industry. Rising energy consumption combined with the environmental impact increases the need for energy saving separation processes. Fortunately complex column networks have the potential for major energy savings estimated to range between 30 and 70% over simple column configuration. In this paper, a computer-aided synthesis method is introduced to synthesize optimal complex column arrangements which encode the cost and states of global solutions with minimum user input. A robust feasibility criterion helps select design and operating conditions for the entire network that can be realized in practice. Our method builds on thermodynamic transformations entitled temperature collocation which provides crucial advantages to determine the operating conditions, structure, and size of the separation network for achieving the desired product cuts. The computational approach guarantees realizable column profiles, which can be validated with industrially accepted simulation software such as Aspen Hysys and Aspen Plus. (C) 2010 Elsevier Ltd. All rights reserved.