Reduction in energy consumption is an important task in process industry. The basic idea of heat exchanger network (HEN) is to cool hot streams using cold streams and to heat cold streams using hot streams. Hence, synthesis and optimization of HEN is a main tool for improving heat recovery. This article introduces a new strategy for HEN optimization using differential evolution algorithm. The proposed method considers splitting stream at the pinch point, to minimize the total cost of the network. Primarily, the minimum approach temperature value is determined through super-targeting. Then, differential evolution is employed to specify the heat load of heat exchangers and splitting streams. The HEN structure obtained in this work has better economics and illustrates the better performance by this approach. © 2014 Trans Tech Publications, Switzerland.